PHISICC - Paper Health Information Systems for health care

Frontline health workers in low- and middle-income countries (LMIC) are increasingly asked by governments and donors to collect more and more data on their activities and to report them to higher levels of the system, compromising their dedication to health care service delivery and overloading them with routine health information systems (HIS) demands [1, 2]. More recently, digital tools have come into play and been promoted by international organisations [3]. While the potential gains brought by digital systems are enormous, they can also worsen this situation as they can easily cause data proliferation and redundancy [4]. Furthermore, in many areas the infrastructure and services to support digital tools are infeasible at present and, in some cases, in the foreseeable future, supporting the approach of hybrid systems [5].

For decades, health information systems for primary health care health facilities have been conceived under the paradigms of “data managers”; hence the proliferation of tools (paper or digital) in tabular forms, disconnected from the processes of care and disregarding the minimal usability features that would encourage their use. Not surprisingly, the generic experience of health workers is that the data they collect is another imposed administrative requirement rather than the basis to inform their health care activities and to improve the measurement of their performance. The result of this situation is often that data is useless for the frontline health workers that collect them and hardly used by the health authorities that request them.

These problems affect mostly LMIC, because of the following factors: (a) donor pressure to quantitatively show the results of their actions leading to an artificial inflation of indicators to measure; (b) limited capacity of the governmental sector to determine their own priorities and approaches (precisely linked to external donor pressure); (c) limited infrastructure and services to implement expensive and highly effective solutions; (d) limited availability and capacity of the health workforce as a result of structural reforms and the LMIC economic context; and (e) the remoteness of health services in rural areas, where a large proportion of the population lives.

Although these issues equally affect urban and rural areas, remote health services are more vulnerable to these issues. More than 3 billion people are living in rural areas in the less developed regions and almost 700 million are living in rural areas in Sub-Saharan Africa [6]. In terms of health workforce, while the number of medical doctors per 10,000 people in Europe is 37, in Africa it is hardly 2, 18 times less [7]. Not surprisingly, it is in those countries where community health workers are more prevalent, with numbers as high as a million in China or 250,000 in the Philippines. These figures can help us to visualise the extraordinary large number of people who rely on very basic health care, delivered by health workers with very limited training and skills. In this context, it is not an exaggeration to suggest that any issue that does not explicitly support health workers' tasks can easily become a barrier. We are afraid that the whole set of data processes in these settings has become a barrier to health care and a strong wake-up call towards the international health community is needed to push for a radical change - to continue doing things as usual is in our view inefficient, hampering quality healthcare, and violating the norms of equity and decolonising global health principles[8].

Our PHISICC solution addressed several of those factors contributing to the problem. First, the development, deployment and testing of our PHISICC solution is based on a strong partnership with Ministries of Health (MOH) and research partners in the South. The partnership with MOH allowed us to ensure that the PHISICC approach would be jointly developed and that there would be no interference from external actors that would interfere in our user-centeredness approach. Second, we tackled the following: (a) optimisation of health care processes, re-designing the information system to fit clinical care and public health needs; (b) simplifying the data processes and reducing the amount of data to collect, to liberate health workers' time for patients’ care; (c) tackle the usability issues of the information tools to improve health workers’ experience with data collection and use.

[1] Siyam, A., Ir, P., York, D. et al. The burden of recording and reporting health data in primary health care facilities in five low- and lower-middle income countries. BMC Health Serv Res 21, 691 (2021). https://doi.org/10.1186/s12913-021-06652-5.

[2] Gimbel S, Chilundo B, Kenworthy N, Inguane C, Citrin D, Chapman R, Sherr K, Pfeiffer J.  Donor data vacuuming. Medicine Anthropology Theory, 5(2) (2018)  https://doi.org/10.17157/mat.5.2.537.

[3] WHO guideline: recommendations on digital interventions for health system strengthening. Geneva: World Health Organization; 2019. Licence: CC BY-NC-SA 3.0 IGO.

[4] Odendaal WA, Anstey Watkins J, Leon N, Goudge J, Griffiths F, Tomlinson M, Daniels K. Health workers’ perceptions and experiences of using mHealth technologies to deliver primary healthcare services: a qualitative evidence synthesis. Cochrane Database of Systematic Reviews 2020, Issue 3. Art. No.: CD011942. DOI: 10.1002/14651858.CD011942.pub2. Accedida el 01 de abril de 2022.

[5] Kumar P, Sammut SM, Madan JJ, et alDigital ≠ paperless: novel interfaces needed to address global health challenges. BMJ Global Health 2021;6:e005780.

[6] United Nations. Department of Economic and Social Affairs Population Dynamics. World Urbanization Prospects 2018. https://population.un.org/wup/DataQuery/ [accessed 05/08/2022]

[7] World Health Organization. The Global Health Observatory. Global Health Workforce statistics database. https://www.who.int/data/gho/data/themes/topics/health-workforce [accessed 05/08/2022]

[8] Khan M, Abimbola S, Aloudat T, Capobianco E, Hawkes S, Rahman-Shepherd A. Decolonising global health in 2021: a roadmap to move from rhetoric to reform. BMJ Glob Health. 2021 Mar;6(3):e005604. doi: 10.1136/bmjgh-2021-005604. PMID: 33758016; PMCID: PMC7993212.

Primary health care performance measurement has to be supported by primary health care performance improvements. It is only by providing interventions to improve performance that health workers can be held accountable about their own performance. Our solution focuses on empowering health workers to make better clinical and public health decisions. 

The PHISICC tools. The PHISICC suite is a totally newly designed information system fitted for remote, rural primary health care facilities, with two components: (i) a set of register books to collect patients information and (ii) a set of tally and reporting sheets to aggregate patients’ information to be sent to higher levels of the health system [1]. Each tool (both for recording and reporting) is specific for a health care area, including the following health care areas: antenatal care, delivery, postnatal care, vaccination, sick child, outpatient consultations, tuberculosis care, HIV care and referral. In total: nine register books and their corresponding tally and reporting sheets. These tools substitute, rather than supplement, the existing register books used in health facilities. Other registers, not linked to patients’ care, are also removed from the system to further reduce the overall load of data collection on health workers.

As sample of PHISICC tools can be found here (and old tools, here).

Technological support and human-centred design approach. In nutshell, the technological approach for our solution consists of  paper +  Human Centred Design + Knowledge, in the form of clinical and public health guidelines. We understand that these aspects configure a technological approach in equal terms as other technologies that rely on sophisticated hardware. Our methods, as described below, were systematic, transparent and informed by the current state of knowledge and could have been equally applied to more sophisticated devices. We argue that our approach was technologically sound, because it took into account the context, the setting, the users, their goals and the availability of resources at present and in the near future. We unfold these issues in the following lines.

In agreement with partners and the MOH, it was decided that the technology to support the PHISICC tools would be paper, given the extremely limited IT capacities in remote, rural areas. It was also decided to print the tools in black and white, with grey tones, to reduce costs and to keep a reasonable printing quality particularly when locally printed. Register books have a DIN-A3 size, differently from current books that tend to be so large that they may reach beyond the limits of the consultation desk once opened. In this way, we ensured that there was enough space to record relevant information and maintaining at the same time the usability and portability of the registers. With this approach, we ensured the adequacy of the tools to the current context where they were meant to be used and mid-term sustainability.

The whole process has been carried out using a human-centred design (HCD) approach: from the understanding of health workers needs and behaviours up to the design of the tools. Professional designers have organised numerous iterations with rural health workers in order to obtain the best possible design, within the technological parameters and respecting the clinical and public health guidelines. The HCD approach has been instrumental and has unveiled problems and opportunity windows that would have remained undetected otherwise. It also allowed us to test and implement different design concepts. For example, we used a clinical process approach for some health care areas (e.g. antenatal care, sick child) and a timeline approach for vaccination (see below for further details). The level of effort was so heavy that for some of the registers we had more than one hundred versions.

It should be noted that the methodological approaches to develop the PHISICC tools were not specific to the technology to be used; i.e. we would have followed a similar thinking, rationale and approaches to develop digital tools. We will come back to this point later in the proposal.

Process of health care. When health workers are seeing a patient, they open the corresponding register (as it is customary also with the old system). For example, the antenatal care register for antenatal care consultations or the vaccination register when a child is being vaccinated. Pages contain standard data items to collect from patients at the time of consultation (e.g. name, address, age, temperature, and so on) and space to fill in those items. Instead of having a tabular form (as is universal in current registers), the information in PHISICC registers is visually structured and consistent with the processes of care: all affiliation information is grouped together, then the anamnesis, then space for vital signs and so on. All registers have the same visual language, in a way that learning to navigate one register means being able to navigate all of them. Depending on the complexity of the health care area, the information for one patient may require one or two pages; for other health care areas, several patients can be accommodated on one page.
It is important to note here that in remote, rural areas the same health workers deliver health care for the whole array of conditions, including maternal and child care, curative services and control of chronic conditions. This is so because of dramatic shortages of staff (e.g. it is not rare to find health facilities with a single cadre) and the nature of primary health care, as the first point of contact for health care. Hence, the PHISICC tools had to cover all those health care areas equally if they had any chance to improve the processes of care. We could have focused, for example, on improving the vaccination register only, which would have been much easier. However, we would have added complexity by adding a ‘new’ tool to the other tools that would have remained unchanged, we would have disregarded the basic principles of primary health care and would have missed linking functions between health care areas (e.g. sick child and vaccination).

Decision-making. The core functionality of the tools is their capacity to support clinical and public health decision-making by frontline health workers, even if (or even more when) they have limited skills. 

It is in the interface patients – health workers where quality of care takes place and what can make the difference between life and death. The PHISICC tools are the information support of this interface. Their importance cannot be overemphasised.

Decision-making support is achieved through several mechanisms:

(a) The visual language incorporates a decision support mechanism that alerts health workers about the potential severity of patients. This is achieved through the use of three different visual icons that are attached to certain data items referring to clinical signs: a square means that the observation or clinical signs is likely of no concern; a diamond means that the clinical signs suggest a problem or a disease; a bold diamond means that the patient may have a life threatening conditions that may require urgent action, including referral. For example, in the sick child registers, one of the items is related to pallor. The decision-making icons (square, diamond and bold diamond) alert health workers about the potential severity of the situation and possible diagnoses. Other examples (e.g. malnutrition, diarrhoeal disease) are more complex but working with the same logic and visual language. See figure below.

(b) The registers guide the health workers through the whole clinical process: from anamnesis to clinical exams till diagnosis and treatment; and because the critical items are already printed in the forms, it is much more easy to be systematic in the assessment of patients and to cover all important areas, just by following the form.

(c) Registers have a dedicated space to record diagnoses, treatment and expected outcome. In this way, health workers are ‘forced’ to go through the clinical thinking, from symptoms and signs to diagnosis and treatment. Let’s not forget that the training of some health workers can be really basic and issues which seem obvious are not so.

(d) The case of vaccination was particular because we implemented a design concept based on a timeline. Because one of the main problems of low vaccination uptake is missed opportunities and loss to follow up of children, we have substituted the usual tabula form with rows for children and columns for vaccines, by a form that represents a timeline. The timeline is represented horizontally and ranges from 0 to 23 months; in the headings of the timeline there is space to write the date and age a child comes for vaccination; and then in the same vertical space, the vaccine received can be written down. In this way, just by looking at the timeline it is extremely easy to detect children who have been lost to follow up, children with delayed vaccinations (i.e. long horizontal spaces between vaccines) or children not vaccinated at all. The timeline design also makes it easy for the healthcare worker to see the need for the vaccinations to follow, and can, therefore, easily inform the client when the next vaccination(s) is needed. 
This design also facilitated the use of information by health workers, who can target specific children for home visits or other vaccination catch up measure.

All information in the register books is based on existing clinical and public health protocols and guidelines, including national vaccination schedules. In fact, what the PHISICC tools do is to ‘translate’ those guidelines into operational forms that facilitate the actual use of information by health workers for decision-making in health care.

Reporting to higher levels of the system. Typically, health workers have to prepare monthly reports to send to the next higher level of the system with aggregated figures describing numerous indicators related to their activity, such as the number of patients seen by health care area or numbers of different diagnoses (note that these may add up to a few hundreds of data items!). They normally do this at the end of the month or at the beginning of the next calendar month; they take an empty report template and then browse through each register to count events, aggregate them and populate all the indicators in the form. These tasks take hours and often a few days of work for one of the staff in the health facility. Taking into account that remote, rural health facilities may have very few staff (often less than five and only one in many places) the monthly reporting tasks take a big toll on their overall capacity to care for patients.

With the PHISICC tools, we have created a tally sheet for each of the nine health care areas. The tally sheets contain all indicators to be reported to the higher levels of the system. For each indicator, there are several rows of small ovals (to tick) and the rightmost third of the tally sheet (separated by a perforated line) contains boxes to write in the counting of the ticked oval for each indicators. At the end of each consultation or clinical events, health workers tick the corresponding ovals. For example, after a normal, first antenatal consultation in a 23 years old woman, they would tick one oval in each of the following antenatal care tally sheet indicators: ‘first consultation’, ‘age 20 to 24’, ‘normal pregnancy’. This tallying may take them a few seconds after each consultation; the time required to say good-bye and to allow the next patient to enter the consultation room. Then, at the end of the month, there is no need anymore to browse through the registers to count events: all of them are already counted in the tally sheets, taking advantage of the ‘dead’ time between consultations. Health workers only need to count the tick marks in the tally sheets, write down the totals, shred the rightmost part of the tally sheet to send it out to the higher levels of the system. The saving of time and effort is remarkable.

How do the PHISICC tools compare with the regular old tools?

• The PHISICC information tools have been developed with a Human Centred Design approach, prioritising health workers experience in clinical care and data management; while usual information tools are developed with a focus on data collection, far from the clinical and public health experience of frontline health workers, where patient care takes place.
• The PHISICC tools embrace the whole primary health care subsystem, which generated a notable complexity but also unveiled systemic issues that would have remained undetected if we would have focused on a single health care area (e.g. vaccination). This is a key difference to many other innovative solutions (including digital solutions) that focus on a single health care area (e.g. vaccination only), disregarding (and not measuring) the harms that having multiple systems produce in the processes of handling a new IT solution for one health care area amidst old tools for the others.
• The PHISICC tools use the whole potential of design, creating a visual language that increases the amount of significance in the forms used, as opposed to tabular sheets which have no remit to the clinical processes they are meant to support.
• The PHISICC tools are process oriented, supporting the core of good quality health care, which is the appropriateness of health workers decisions in their diagnoses and treatment, as opposed to data items without clinical remit.
• The PHISICC tools split the health care and data management functions, boosting both functions: making information supporting clinical care more available and meaningful and reducing the time required for reporting to higher levels of the systems.
• The PHISICC tools have converted the information system into a quality of care instrument and the bloodstream of health care, instead of having data collection tools that keep inflating the data requirements in remote health facilities while they are hardly used.

[1] Bosch-Capblanch X, O'Donnell D, Krause LK, Auer C, Oyo-Ita A, Samba M, Matsinhe G, Garba AB, Rodríguez D, Zuske M, Njepuome AN, Lee SMM, Ross A, Gajewski S, Muloliwa AM, Yapi RB, Brown DW. Researching, co-creating and testing innovations in paper-based health information systems (PHISICC) to support health workers' decision-making: protocol of a multi-country, transdisciplinary, mixed-methods research programme in three sub-Saharan countries. Health Res Policy Syst. 2021 Aug 11;19(1):112. doi: 10.1186/s12961-021-00768-0. PMID: 34380518; PMCID: PMC8355573.

Our aim is clearly to improve the quality of care for the most vulnerable populations through a solution that improves the performance of health workers in delivering quality of health care. Hence, our solution serves both (i) the general population and (ii) health workers, particularly living in remote, rural areas.

Health care in remote, rural areas, such as in Sub-Saharan Africa, can be extremely precarious due to the lack of services, infrastructures, equipment and manpower [1]. The quality of health services in LMIC, with higher proportion of population living in rural areas, is generally low and it is associated with higher mortality [2]. In remote, rural areas there are no redundant services that can palliate poor quality in designated health centres (e.g. stock-outs may deter patients from accessing services [3]) and often referral mechanisms are hardly developed [4]. Poor early detection of risk situations [5] often leads to harmful health outcomes.

In synthesis, health care in remote, rural areas is supported by an array of very few tools; essentially: the availability and motivation of a few health workers with limited skills, some basic infrastructures, often without electricity and reliable water supplies and a few essential medicines (some in stock out) in a very harsh socio-economic context. In this situation, any further failure of the system may have catastrophic consequences. Hence, the importance to strengthen what is already available together with improving the availability of resources and services. The PHISICC tools aim at securing one of those critical aspects, which relates to health workers performance in good quality of health care: their decision-making capacity in the interface between patients and health workers. That is the place where correct decisions on diagnoses and treatments are likely to solve a health problem and where wrong decisions lead to poor health outcomes and death.

[1] Strasser R, Kam SM, Regalado SM. Rural Health Care Access and Policy in Developing Countries. Annu Rev Public Health. 2016;37:395-412. doi: 10.1146/annurev-publhealth-032315-021507. Epub 2016 Jan 6. PMID: 26735432.

[2] Kruk ME, Gage AD, Joseph NT, Danaei G, García-Saisó S, Salomon JA. Mortality due to low-quality health systems in the universal health coverage era: a systematic analysis of amenable deaths in 137 countries. Lancet. 2018 Nov 17;392(10160):2203-2212. doi: 10.1016/S0140-6736(18)31668-4. Epub 2018 Sep 5. Erratum in: Lancet. 2018 Sep 20;: PMID: 30195398; PMCID: PMC6238021.

[3] Kuwawenaruwa A, Wyss K, Wiedenmayer K, Metta E, Tediosi F. The effects of medicines availability and stock-outs on household's utilization of healthcare services in Dodoma region, Tanzania. Health Policy Plan. 2020 Apr 1;35(3):323-333. doi: 10.1093/heapol/czz173. PMID: 31942625; PMCID: PMC7152726.

[4] Ofosu B, Ofori D, Ntumy M, Asah-Opoku K, Boafor T. Assessing the functionality of an emergency obstetric referral system and continuum of care among public healthcare facilities in a low resource setting: an application of process mapping approach. BMC Health Serv Res. 2021 Apr 29;21(1):402. doi: 10.1186/s12913-021-06402-7. PMID: 33926425; PMCID: PMC8082760.

[5] Jeeana S, Naidoo TD. An appraisal of the referral system and outcome for obstetric patients referred to a tertiary centre. Trop Doct. 2022 Aug 2:494755221116718. doi: 10.1177/00494755221116718. Epub ahead of print. PMID: 35918834.

Our work has been carried out in Côte d’Ivoire, Mozambique and Nigeria. Memoranda of Understanding have been signed between the different partners. The PHISICC core team is composed by:

• Coordination: based at the Swiss Tropical and Public Health Institute, in charge of harmonising methods, ensuring that activities are implemented according to plans, making funding available and reporting to the funding bodies.
• Human Centred Design experts, who work across the three countries in two main areas: understanding the problem and how health workers interact with information tools, patients, co-workers and the environment; and co-creating with health workers the PHISICC solution.
• Researchers in the three countries, composed by a mix a genders, ages and expertise, leading the development, implementation and testing of the PHISICC solution in each country.
• Ministry of Health officials, who have provided overall guidance, administrative clearances and numerous inputs along the whole process of developing and testing the PHISICC solution.

The PHISICC team, composed by public health specialists, health systems experts, human-centred designers, anthropologists, social scientists, and health care providers, with more than 30 core people from six countries, really proved in real terms the potential of transdisciplinary collaboration creating the PHISICC tools, a really first class health information systems solution. Our approach was publicly praised twice by a renowned Professor of Public Policy at the IOE - Social Research Institute (UCL Institute of Education), member of the PHISICC Technical Advisory Group here and here.

Furthermore, numerous stakeholders have contributed with numerous technical inputs and advices; and most importantly, health workers have been an integral part of the whole process, since ideation, up to testing, including co-creation design and implementation. Numerous exchanges during months have allowed the designers to tailor the solution to the needs of health workers in providing health care.

The fact that local researchers are part of the same communities that we are serving, the fact that they have been carrying out research and development activities in those communities, and the extensive exchanges with health workers for months during the co-creative phases of the project, ensured that the ideation, design and implementation of the solution was rooted at all time on the actual reality of those settings and communities.

For example, in order to understand the needs of health workers and the patients they serve, we started by carrying out a characterisation of the health information system: at the outset of the PHISICC project, local researchers and designers visited remote, rural health facilities in the three countries to understand health workers' behaviours, interactions and relations with the health information system. Several techniques were used, including shadowing, focus group discussion, key informant interviews and observations. The output of this activity was a mapping of the health care processes and use of the existing information systems. Further information can be found here (see section: WS3 – Characterisation of the Health Information Systems in Nigeria, Côte d’Ivoire and Mozambique).

During the co-creative stages, and based on the characterisation of the health information systems, together with research evidence gathered to learn from other settings [1], we discussed a series of criteria that the solution should comply with; e.g. user-friendly, with a clear and standard visual language, able to guide decision-making processes, consistent with current clinical and public health guidelines, adapted to the context, affordable and sustainable. These initial prototypes were circulated among partners and stakeholders to gather first inputs and to rule out features that seemed inappropriate. Then, the designer teams travelled to Mozambique to visit rural health facilities on the spot and to engage with health workers into a dialogue to gather their views in relation to different design options. The local teams in Côte d’Ivoire and in Nigeria, with the support of the designers, carried out the same activities. Health workers were approached in several iterations to present them with the materialisation of the options discussed in previous sessions. This process was repeated since designers got consistent feedback from a variety of health workers in the three countries. Then, the whole work consisted in fine-tuning the designs and addressing issues that emerged in different health care areas (e.g. lack of space of certain information or the need to be comprehensive but also keep simplicity). Not surprisingly, health workers were thrilled with the result, since the new PHISICC tools were really their own product.

[1] Zuske MK, Auer C, Oliver S, Eyers J, Bosch-Capblanch X. Framework synthesis to inform the ideation and design of a paper-based health information system (PHISICC). Int J Health Plann Manage. 2022 Jul;37(4):1953-1972. doi: 10.1002/hpm.3487. Epub 2022 Apr 23. PMID: 35460301.

Through the PHISICC experience our perspective was widened to the plight of health workers in peripheral settings. Through interactions with them and in collaboration with designers, we were able to design health information tools that support the health workers in their daily tasks and it led us to the believe that fundamental changes in the way HIS are designed, implemented and used are urgently needed. We have to stop doing business as usual when developoing health information systems. 

Therefore, we plan  to create a multidisciplinary Technical Health Information Systems working group (THIS-WG) to build consensus on the research and implementation parameters of the health information systems of the future, as quality of care, human centred designed instruments that empower health workers to improve the health status of the population. This working group would be composed by Ministry of Health officials, health workers, researchers, designers, multilateral organisations (e.g. WHO) and funders, and would operate during a 12 to 18 months period, to produce a guidance document to support future health information systems developments.

What supports the rationale for the THIS-WG?

(i) With PHISICC, we have demonstrated that a systemic change focusing on health workers and quality of health care (as opposed to mere data collection) is feasible, well appreciated by health workers [1] and with the potential to improve quality of cafe;

(ii) even more importantly, we have outlined a methodological, transdisciplinary approach, human-centred, robust, transparent and systematic that can be applied to the development of other systems and sub-systems, being paper, digital or, more likely, hybrid; and

(iii) our experience should be shared and contrasted with that of other groups working in the same direction and equally interested in making health information systems quality of care instruments.

Precisely because we have produced true innovation, our experience cannot easily fit in existing thinking paradigms or in current structures and organisations. Novelty requires change at multiple levels and we are precisely in this crossroad, where the resistance to change and to conceive alternatives to what the international aid and development community has done for decades is the main barrier to progress.

We would like to ask for the patience of the reader to unfold some of these issues in the following lines.

We, the PHISICC research team, have experienced ourselves the hardship to change our way of approaching the health information systems issues. This is what happened to us: our starting point, back in 2016, was to ‘simply’ operate some design improvements in the existing paper-based tools to improve data quality being produced in health facilities. At the very start, we spent some time working under the ‘usual’ perspective: health workers have to provide data to the higher levels of the system, in order to justify funds, development aid and to support planning of services. However, we realised that (i) health workers experienced data collection and management as imposed tasks to them; (ii) health information systems in high income countries are much less demanding on health workers providing care (although they also complain); and more importantly (iii) we did not really grasp how vulnerable communities accessing health services could actually experience a real benefit from a change in the information system.

This caused us to go back to the design table and rethink as a team. Were faced two rather simple but critical questions, largely posed by designers: (i) what is the main mission of health workers?; and (ii) what do health information systems have to do with this mission? We realised the health workers’ mission is to take care of the population and patients and we had to hypothesise that the health information system would support this mission (among other functions, such as reporting to higher levels). The key to solve the puzzle was considering ‘decision-making’ for reaching diagnoses and prescribing treatments, as the core competence of health workers and what needed to be supported in the first instance by any information system. This brought us (the team) to the paradigm shift: from health information systems focusing on collecting and reporting data, to health information systems as quality of care instruments, supporting health workers in the interface with patients, where decisions about life and death are made.

How is the ‘paradigm shift’ barrier concretised? There are two main barriers:

1) despite the successes of PHISICC (and other experiences), some Ministries of Health staff, at national and sub-national levels, even some health workers, experience a strong reluctance to system changes, feeling that changes may bring losses of data and, hence, punishments in the governmental sector, which is heavily hierarchical and bureaucratic;

2) digital tools are occupying the research on and implementation arenas with little attention to the systemic consequences of these tools and to the need to provide more than anecdotal evidence on their impact on health systems and health outcomes.

The THIS-WG would deal with all those issues to support and inform future developments on health information systems.

Clarification note. We do not see any debate, conflict or contraposition between paper and digital tools. We worked on paper because that was our remit in the specific context of remote, rural areas and our agreement with Ministries of Health. We strongly believe that health information systems will have to be hybrid, multifunctional and level of care-specific, as they are in high income countries. What we advocate for, in a nutshell, is that any development on health information systems has to be (i) human-centred and (ii) tested using robust study designs. For example, testing a digital tool circumscribed to a single health care area (e.g. an electronic vaccination record) in a few health facilities may work only because the intervention is relatively small and manageable, it does not bring changes in the process of care and it is tested without a suitable comparison or counter-factual. We should not forget that despite some recommendations in line of digital tools [2], the research evidence supporting digital tools in health care is not even minimally conclusive [3].

[1]  Nnette Ekpenyong, Kathrin Heitz Tokpa, Ogonna Nwankwo, David O'Donnell, Damaris Rodriguez Franco,  Salimata Berté, Simplice Amani Kouassi, Glory Eteng,  Veronica Undelikwo, Christian Auer, Gouzan Bernard Guessan, Angela Oyo-Ita, and  Xavier Bosch-Capblanch. Using and improving the PHISICC paper-based tools in the Health Facility Laboratories: examples of Human Centred Design taking Systems Thinking into practice, in Côte d’Ivoire and Nigeria https://www.frontiersin.org/articles/10.3389/fpubh.2022.916397/abstract [full text in press - accessed 07/08/2022]

[2] WHO guideline. Recommendations on digital interventions for health systems strengthening. World Health Organizations 2019.

[3] Agarwal S, Glenton C, Tamrat T, Henschke N, Maayan N, Fønhus MS, Mehl GL, Lewin S. Decision-support tools via mobile devices to improve quality of care in primary healthcare settings. Cochrane Database Syst Rev. 2021 Jul 27;7(7):CD012944. doi: 10.1002/14651858.CD012944.pub2. PMID: 34314020; PMCID: PMC8406991.

Our PHISICC solution, as outlined above, has two main contributions: (i) as a suite of newly designed paper tools to empower health workers decision-making; and (ii) as a methodological approach to support health information systems reform. The innovation character of PHISICC consists in considering health information systems as quality of care instruments, as opposed to mere data collection tools.

In terms of the PHISICC tools:

• in a context largely dominated by digital technologies, the use of paper to ideate and produce a health information systems is indeed innovative; we took a strategic decision with the Ministries of Health, researchers and designers, to produce a solution that could be readily available, adapted to the context and sustainable;
• the tools were consistent with the processes of care and were designed to guide these processes, instead of tabular formats that provide no added value in terms of health care tasks;
• the processes of care were supported by a visual language, clear, standard across health care areas and providing additional meanings to the visual elements;
• we created two data sub-systems: data for clinical management and data for reporting to higher levels of the system. In this way, we could get the most of each sub-system instead of compromising one in order to achieve the full functionality of the other. We connected both sub-systems through the tally sheets, which significantly reduced the workload of health workers when doing reporting at the end of the month.

In terms of the PHISICC methodological approach:

• we successfully implemented a transdisciplinary approach, combining and creating synergies from such diverse disciplines as public health, anthropology, social sciences, human centred design and health systems;
• we focused on quality of health care and on empowering health workers to make better clinical and public health decisions, through a Human Centred Design approach;
• we embraced the whole primary health care system and not only a small part of it or a single health care area; this approach brought up challenges that would have remained unveiled; challenges that we addressed through design artefacts (e.g. how to build references within several health care areas in a way that the paper form reminds health workers that a sick child has to be considered for his or her vaccination status as well);
• we used robust experimental study designs [1] to test the effects of the PHISICC tools. It is worth noting that we used cluster randomised controlled trials (RCT) in the three countries to test the PHISICC solution. RCT are the gold standard to test interventions because they have the lowest risk of bias. They are also more demanding in terms of effort and technical capacity [2]. We were decided to carry out the best possible studies even if the study settings were extremely challenging in remote, rural areas in Côte d’Ivoire, Mozambique and Nigeria. This is also important because development of guidelines (e.g. by WHO [3]) strongly relies on evidence coming from systematic reviews of RCT. This entails, in practical terms, that RCT tend to make it into guidelines and policy, while observational studies, more common but more biased, don’t. Clearly, the methodological quality gone into the development PHISICC deserved the gold standard of study designs, also for testing it.

We believe that these innovations, in combination with the experiences produced in other initiatives to improve health information systems, and considering the momentum of PHISICC, can bring partners and stakeholders together in order to rethink and conceive the health information system of the future. Among the wider international health and development community, we do have enough evidence and experience to stop doing things as usual and to radically improve how information systems are conceived and implemented to serve the most vulnerable populations. We expect that the THIS-WG, with the participation of relevant actors, such as WHO, academia, ministries of health, designers and health workers should be able to scrutinise the available body of knowledge and methods to develop health information systems, identify and ban bad practices, and propose new standards for health information systems research and implementation. We have done it in PHISICC and we expect that this can be done with the wider community for greater impact.

The health information systems of the future will be quality of care instruments, will empower health workers, will serve the most vulnerable populations and will be supported by hybrid technologies.

[1] Bosch-Capblanch X, Oyo-Ita A, Muloliwa AM, Yapi RB, Auer C, Samba M, Gajewski S, Ross A, Krause LK, Ekpenyong N, Nwankwo O, Njepuome AN, Lee SM, Sacarlal J, Madede T, Berté S, Matsinhe G, Garba AB, Brown DW. Does an innovative paper-based health information system (PHISICC) improve data quality and use in primary healthcare? Protocol of a multicountry, cluster randomised controlled trial in sub-Saharan African rural settings. BMJ Open. 2021 Jul 29;11(7):e051823. doi: 10.1136/bmjopen-2021-051823. PMID: 34326056; PMCID: PMC8323359.

[2] Wolfenden L, Foy R, Presseau J, Grimshaw JM, Ivers NM, Powell BJ, Taljaard M, Wiggers J, Sutherland R, Nathan N, Williams CM, Kingsland M, Milat A, Hodder RK, Yoong SL. Designing and undertaking randomised implementation trials: guide for researchers. BMJ. 2021 Jan 18;372:m3721. doi: 10.1136/bmj.m3721. PMID: 33461967; PMCID: PMC7812444.

[3] WHO. WHO Handbook for Guideline Development. 2nd Editions. WHO 2014.

• Our long-term impact goal is to improve the quality of health care of the most vulnerable communities, particularly those living in remote, rural areas in LMIC.
• Our mid-term goal is to transform current health information systems, particularly in LMIC, towards systems that empower health workers to improve their performance in clinical care and public health, leading to better quality of care.
• Our short-term goal is to produce a position paper on methods and approaches to transform health information systems, from data collection tools to quality of care and performance improvement instruments.

We expect to form a Technical Health Information Systems working group (THIS-WG), composed by ministries of health members from LMIC, researchers, health workers and designers, from different institutions and constituencies. The mandate of the THIS-WG will be to scrutinise the current evidence and approaches related to health information systems research and implementation. Based on this work, the THIS-WG will propose best practices and suggest scenarios on how to transform current systems into instruments to improve the quality of care and the population's health status. We expect that the THIS-WG could complete its work in a 12 to 18 months period and produce a position paper on the matter (short-term goal).

In parallel, the THIS-WG will identify countries that are willing to embrace such transformation, with particular consideration to Côte d’Ivoire, Mozambique and Nigeria. These countries will be supported into making the position paper operational and be the pioneers in substituting obsolete systems with information systems based on a quality of care vision and likely hybrid in terms of technological support (mid-term goal).

In the long-term, we expect that quality of care will be improved and that this will be translated into better health outcomes for patients but also in community based interventions. The position paper would include guides on how to measure the impact of health information systems transformations (long-term goal).

We propose the following set of indicators. Because the innovative health information system focus on quality of care and health workers performance, they bring more accountability as well, and allow to measure primary health care performance at the core of its mission, which is to provide health care at the first point of contact of the communities.

• Process indicators
  • Partners and stakeholders identified to form the THIS-WG (year 1)
  • Terms of Reference of the THIS-WG agreed (year 1)
  • THIS-WG constituted and operational (year 1)
  • Position paper produced (year 2)
  • Countries to engage into Health Information Systems transformation identified (year 2)
  • Transformation activities started (year 2)
  • Transformation implemented (years 2 and 3)
• Output indicators (most of them estimated for PHISICC and proposed for the THIS-WG in years 3 to 5, as well)
  • adherence to clinical guidelines by health workers
  • adequacy between diagnoses and treatments by health workers
  • stock-outs of essential commodities
  • health workers satisfaction with the processes of care
  • patient and public satisfaction with the provision of care
  • ratio of patient care / management time used by health workers
• Outcome indicators (most of them estimated for PHISICC and proposed for the THIS-WG in years 3 to 5, as well)
  • Quality of care improvements (in process of being analysed for the PHISICC tools)
  • Rate of identification of pregnancies at risk
  • Vaccination coverage rates increase
  • Effectiveness of the referral system

In the following lines we outline the inputs that will be organised in THIS-WG processes, the activity of which will produce the output of the position paper, as the basis for health information systems transformation, that will improve the quality of care and health outcomes.

Inputs

1) PHISICC and other projects achievements. We are at present finalising all the analyses that will describe the processes and effects of the PHISICC tools in the three countries, using both qualitative and quantitative approaches. Preliminary analyses suggest that PHISICC have been highly appreciated by health workers and even patients and there have been improvements in quality of care indicators (see preliminary qualitative results in Ekpereonne et al [1]).

2) PHISICC and other projects dissemination. The dissemination of PHISICC results has already started and will raise the interest of research partners themselves, stakeholders in the three countries and the wider research community. Apart from scientific papers, we are already disseminating our findings in conferences, such as in an organised session at the Seventh Global Symposium on Health Systems Research, Bogota 2022 [2].

3) Interest on health information systems transformation. We count on the strong interest of PHISICC partners, from Ministries of Health and research teams in Côte d’Ivoire, Mozambique and Nigeria. Besides, we have already started exploring the interest of stakeholders in considering the issue of health information systems transformation.

4) Availability of local and global evidence. We consider that the amount of evidence related to health information systems research and implementation is large enough to inform the production of the position paper by the THIS-WG (see for example the paper by Zuske et al [2]).

We expect those four inputs can generate enough momentum to constitute the Technical Health Information Systems working group (THIS-WG).

Processes

1) THIS-WG creation. The momentum created by the issues outlined above will allow the creation of the THIS-WG. This is the cornerstone of our activity, since it takes the existing level of knowledge (including PHISICC learnings) into the operational field, setting up the parameters of health information systems transformation.

2) THIS-WG reputation. Based on our past experience and current relations and networks, we are confident that a THIS-WG can be formed with competent and committed individuals and that it can effectively and efficiently produce the output of the health information systems position paper. The fact that we have been successfully working with most of these partners and stakeholders over the last six years in PHISICC supports our statements.

3) THIS-WG modus operandi. We will ensure that THIS-WG members are kept updated at all time, facilitating the communication means and a sensible mix of online and in person meetings. We will also carefully assess the workload of its members and their capacity to make contributions. The THIS-WG will be accompanied by Human Centred Design experts, who have a proven track record of being able to articulate different perspectives towards a common goal.

4) THIS-WG outputs. The team, partners and stakeholders have the mix of expertise to ensure the management of the working group through an efficient secretariat (e.g. setting up of agendas and writing of minutes) and to generate discussions that can bring clarity to the issues raised.

THIS-WG activities will result in the production of the health information systems position paper.

Outputs

1) Position paper standards. The position paper will outline the standards to carry out research and implementation of newly conceived health information systems, drawing from research evidence and the experience of various research groups.

2) Countries identification. In parallel, the working group will identify countries willing to engage into such transformation. This should be feasible based on (i) the credibility of the THIS-WG and its outputs; (ii) the experience of PHISICC and other initiatives aiming at transforming health information systems; (iii) the viability of the processes outlined in the position paper; and (iv) the efforts of the THIS-WG to identify funding sources to support these transformations.

3) Countries engagement. Because the THIS-WG will count with influential individuals at several levels of the international health arena, we expect that at least two or three countries will be willing to engage into health information systems transformation. We count on the PHISICC experience that clearly demonstrates that there are Ministries of Health willing to engage into transformative actions.

Outcomes

1) Quality of care improvements. Since the newly conceived health information systems are designed as quality of care instruments, and based on the preliminary evidence from PHISICC, it is reasonable to expect that quality of care improvements will take place.

2) Health workers performance. The new health information systems, by design, will empower health workers in their capacities to make clinical and public health decisions. In turn, this will make performance measurement more meaningful, by (i) focusing on the core competence of health workers and (ii) implementing well accepted and tested quality of care measurement tools (e.g. audit and feed-back [4]).

3) Health outcomes. Improvements in quality have been associated with better health outcomes (see for example Carinci et al [5]).

Our theory of change, though, is not exempt of risks.

1) We believe that the main risk is related to the capacity and willingness from stakeholders that focus on data management issues, rather than on quality of care, to change the thinking paradigm, as we propose. To this end, we will carefully structure the THIS-WG in a way that nobody feels excluded and that all perspective can be openly discussed. We count into incorporating Human Centred Design experts for mentoring the THIS-WG, to ensure that the group moves forward towards the common goal of transforming health information systems.

2) The heavy workload of THIS-WG members may jeopardise their effective participation. We will mitigate this risk by creating agile communication channels, synthesising and summarising the evidence materials to be used in the group and creating synergies. We will focus our efforts in a few critical events that will required to make strategic decisions on how to transform health information systems.

[1]  Nnette Ekpenyong, Kathrin Heitz Tokpa, Ogonna Nwankwo, David O'Donnell, Damaris Rodriguez Franco,  Salimata Berté, Simplice Amani Kouassi, Glory Eteng,  Veronica Undelikwo, Christian Auer, Gouzan Bernard Guessan, Angela Oyo-Ita, and  Xavier Bosch-Capblanch. Using and improving the PHISICC paper-based tools in the Health Facility Laboratories: examples of Human Centred Design taking Systems Thinking into practice, in Côte d’Ivoire and Nigeria https://www.frontiersin.org/articles/10.3389/fpubh.2022.916397/abstract [full text in press - accessed 07/08/2022]

[2] Research, Rethink and Reform: the PHISICC approach to ideate the Health Information Systems of the future. Organized session https://healthsystemsresearch.org/hsr2022/program/ [accessed 08/07/2022]

[3] Zuske MK, Auer C, Oliver S, Eyers J, Bosch-Capblanch X. Framework synthesis to inform the ideation and design of a paper-based health information system (PHISICC). Int J Health Plann Manage. 2022 Jul;37(4):1953-1972. doi: 10.1002/hpm.3487. Epub 2022 Apr 23. PMID: 35460301.

[4] Ivers N, Jamtvedt G, Flottorp S, Young JM, Odgaard-Jensen J, French SD, O'Brien MA, Johansen M, Grimshaw J, Oxman AD. Audit and feedback: effects on professional practice and healthcare outcomes. Cochrane Database Syst Rev. 2012 Jun 13;(6):CD000259. doi: 10.1002/14651858.CD000259.pub3. PMID: 22696318.

[5] F. Carinci, K. Van Gool, J. Mainz, J. Veillard, E. C. Pichora, J. M. Januel, I. Arispe, S. M. Kim, N.S. Klazinga, on Behalf of The OECD Health Care Quality Indicators Expert Group, M. Haelterman, P. Meeus, J. Lacroix, J. Cenek, C.R. Barsøe, K. Grau, L. Rooväli, P. Hämaläinen, V. Garcia, C. Grenier, B. Le Cossec, M. Marbach, C. Scheidt-Nave, D. Mulholland, A. Ekka-Zohar, T. Kumakawa, E. Okamoto, E.H. Byeon, K.H. Kim, C.S. Park, J. Lepiksone, F. Berthet, C. Margue, M. Van Den Berg, A.K. Lindahl, H. Narbuvold, E. Dudzik-Urbaniak, P. Boto, E.K. Lim, W.Y. Mok, R. Pribakovic, M. A. Gogorcena, M. Aggestam, M. Köster, M. Lawrence, M. Langenegger, K. Fehst, S. Yilmaz, K. Everard, V. Raleigh, on Behalf of The OECD Health Care Quality Indicators Expert Group, Towards actionable international comparisons of health system performance: expert revision of the OECD framework and quality indicators, International Journal for Quality in Health Care, Volume 27, Issue 2, April 2015, Pages 137–146, https://doi.org/10.1093/intqhc...

Our technological approach to develop the PHISICC tools and to produce the position paper on health information systems transformation by the THIS-WG consists of: (1) knowledge; (2) Human Centred Design; and (3) paper.

We expect that the position paper will consider the whole array of issues related to health information systems. Areas to cover include, for example: primary health care, hospital-based care, disease surveillance, logistic health information systems, human resources information systems, health facilities master lists, procurement and supplies, health care financing, universal health coverage, preventive measures or community health, to mention a few. The health information system and sub-systems cannot be fragmented and isolated for each one of the areas just listed. Issues such as data security, interoperability (e.g. with DHIS2 [1]), robustness or interfaces, have to be considered within the overall systems perspective and human centred approach. This will bring into the table the whole array of technologies, from paper to digital, from statistics to artificial intelligence, from smart phones to servers.

In the position paper, the THIS-WG will establish scenarios and describe health sub-systems and information requirements and will map specific technological options onto them. It will also suggest how countries may advance towards the scenarios in order to make their health information systems quality of care instruments, improving the performance of health workers and health manager as well.

[1] Dehnavieh R, Haghdoost A, Khosravi A, Hoseinabadi F, Rahimi H, Poursheikhali A, Khajehpour N, Khajeh Z, Mirshekari N, Hasani M, Radmerikhi S, Haghighi H, Mehrolhassani MH, Kazemi E, Aghamohamadi S. The District Health Information System (DHIS2): A literature review and meta-synthesis of its strengths and operational challenges based on the experiences of 11 countries. Health Inf Manag. 2019 May;48(2):62-75. doi: 10.1177/1833358318777713. Epub 2018 Jun 13. PMID: 29898604.

Because our solution in the THIS-WG position paper will involve the whole array of (or the main) information sub-systems, data collection may come from different sources, including, for example: patients’ data produced in clinical care and public health events; data on supplies and stocks or data on human resources capacity, availability and turnover.

All these data are produced within the health system. In small primary health care centres, it is the same health worker who provides care and collects all data requested; in larger health centres, hospitals and health services managerial entities, there may be dedicated staff who collect data.

Inherent to our solution is the attempt to simplify data collection, reduce data items that have no clinical and public health utility, and adapt the processes in a way that data collection does not interfere with patients’ care.

Diversity. Our PHISICC partnership is by design inclusive of a large array of diversity, in terms of age (young and older experts), gender, level of expertise, ethnicity, country of origin and any other feature. Because the nature of our work is transdisciplinary, we actively search for this diversity in our partners. We expect that the THIS-WG will operate on a similar basis.

Equity. The PHISIC project made an explicit choice to focus on the most vulnerable communities in remote, rural areas and on the health institutions serving them. Our solution has been explicitly fit to those contexts in terms of design, clinical priorities and public health activities supported by the information system. We expect that the THIS-WG will operate on a similar basis.

Inclusion. The PHSICC and THIS-WG aspire to draw technological options and approaches from every possible field, with the only conditions that they are fit for purpose and supported by a reasonable body of evidence. In PHISICC we involved a whole array of stakeholders in the different phases of the project, convinced that ownership of the PHISICC tools had to expand from patients’ to health care managers at central level. This was not only an ethical requirement, but also one of the most fruitful experiences in PHISICC and a key contribution to make it possible. We expect that the THIS-WG will operate on a similar basis.

Furthermore, we operate under the Swiss principles of cooperation [1].

[1] Schweizerische Eidgenossenschaft. International cooperation. Principles of cooperation. https://www.eda.admin.ch/deza/... [accessed 08/08/2022].

As suggested, in the following lines, we are using the Social Business Model Canvas.

Key resources. The THIS-WG is the key resource. It will operate as an expert, multi-disciplinary body to produce the health information systems transformation position paper. For the actual phase of transforming health information systems in selected countries, funding will come from Ministries of Health running budgets and potentially from external donors.

Partners and key stakeholders. Multilateral agencies (e.g. WHO), Ministries of Health in selected countries, researchers in several disciplines, health workers and professional designers.

Key activities. Analysis and synthesis of the evidence on health information systems transformation; framework analysis of health systems dimensions and health care levels; production of scenarios integrating dimensions, levels and information requirements; considerations on technological options to transform health information system into quality of care instruments; writing and publication of the position paper.

Type of intervention. Constitution of a working group. Online meetings. Key face-to-face meetings. Production of publication.

Beneficiaries. Populations accessing health services, particularly in LMIC; frontline health workers.

Customer. Ministries of Health who have to be in the driving seat, supported by funding agencies.

Channels. Selected countries will be identified during the THIS-WG sessions. The position paper will be published and disseminated in the literature and conferences. Partners will promote the approaches in the position paper through their own networks.

User value proposition. A newly designed health information system that is conceived as a quality of care instrument, it is centred on the health worker’s experience and is supported by hybrid technologies, will improve the quality of health care, reduce workload and improve health outcomes in the populations accessing health services.

Impact measures. Examples include: increase in vaccination coverage, increase in detection of pregnancies at risk, more effective referrals from primary health care and, eventually, reduction of mortality.

Customers value proposition. A newly designed health information system that is conceived as a quality of care instrument, it is centred in the health workers experience and is supported by hybrid technologies, will better support decision-making at all levels, will rationalise health care processes and will improve the efficiency of the system. As a result, the performance of health sub-systems will be measured more easily and with greater accuracy, allowing a better planning and management of existing resources.

Cost structure. The main cost will take place at the design stage, although these costs will be reduced by the fact that the PHISICC experience can already provide solutions to some of the methodological challenges (e.g. how to articulate design requirements and clinical guidelines). These costs will soon fade out as new methods are well established and implemented. The other costs refer to the specific technology mix decided by each country, depending on the levels of the system to target and the socio-economic context.

Surplus. There will not be surplus as such. We expect though to gain efficiencies in the delivery of health care.

Revenue. There is no planned revenue. We expect though the gain efficiencies in the delivery of health care.

Health Information Systems are part of the core structures and functions of Ministries of Health. They are funded through the core funding of Ministries although in some cases may have the support of external donors, particularly for the production of paper tools and the introduction of digital technologies.

We expect that the THIS-WG will consider sustainability issues in the scenarios and options proposed based on the existing evidence on investment and maintenance costs of services, equipment and supplies. We are currently carrying out an economic analysis of PHISICC costs.

e Swiss TPH operates on a non-for profit basis and the products and services that we develop are the result of our research and technical assistance, funded through research grants and philanthropies. Our income also comes in a small proportion from governmental sources and from teaching activities.