1
Paper to be presented at
The Nordic Welfare Research Conference.
Towards Resilient Nordic Welfare States.
Challenges, Responses and Consequences
14-15 March 2019, University of Helsinki, Finland
By
Carin Björngren Cuadra
Associate professor in Social work
Malmö University, Sweden
2
Technology-dependent documentation and social
redundancy – a scenario-based analysis of an IT
failure in the social services
In this article, I examine what failures or breakdowns in information technologies mean for the operations within the social services. I do this based on a special interest in the documentation of the core activities. My starting point is that documentation has a particular status within the social services, which are, according to the Social Services Act (SFS 2001:453), obliged to document the handling of matters related to individuals. This obligation encompasses both the investigation and the implementation of decisions regarding income support and other investigations (Fahlberg & Larsson 2016). Besides the required
documentation, the staff, according to my experience, use forms of documentation chosen by themselves as memory tools. This can be in the shape of informal memos, notes taken on the mobile phone, or a sticky note on the office door.
Documentation is performed routinely in special operational programmes and in standardised methods for treatment work (e.g., ASI, Addiction Severity Index) with the support of computerised documenting practices. Staff in managerial positions document by means of special programmes for scheduling and staff planning, budget work, quality work, work environment management and incident reporting. Thus, documentation is to a great extent performed by means of
information technology, IT. IT enables not only computerised support when writing, in that a computer has replaced the typewriter and saves what you write on a central server; it also makes it possible to easily share documentation with colleagues through a joint authorisation to access certain operational programmes. The same technology enables communication, both within and outside the
organisation through, for example, intranets, e-mail, voice over IP and conversations via platforms such as Skype.
The social services are thus dependent on the functioning of IT in order to perform everyday documentation and communication. According to a head of unit in elderly care who participated in my investigation, “much depends on computers and IT.” This is the case not merely with regard to record keeping and work in
3
operational programmes, but also with regard to other work tasks. For example, an assistant nurse in elderly care may have the buzz code of a user in their mobile phone, whose technology is connected to both the safety alarm and the door lock of the user. The questions that arise in this context are: What happens in case of an IT failure? Is it possible to maintain the social work?
This type of technology dependence is an important dimension of the
complexity of modern society (Johansson & Hassel 2016). The technology offers new possibilities. The municipal home care nursing may, for example, offer advanced health care in the home, just as the social services may provide web-based social counselling and documentation by means of a mobile phone during home visits. However, the technology also entails an increased vulnerability (ibid.).
Different forms of technologies have connections and mutual dependencies, which means that failures may, in individual cases, result in incalculable and serious knock-on effects in several stages (ibid.). In Sweden, the municipalities have a responsibility1 to prepare an analysis of what serious events2 may occur and to take measures to reduce the vulnerabilities of the operations – such as those of the social services – in case of such events. This is part of the crisis
preparedness3 of the municipality which, among other things, encompasses
so-called risk and vulnerability analyses (Swedish Civil Contingencies Agency [MSB] 2011). I will return to those analyses, but I want to point out already at this stage that working with information security covers the protection of the
confidentiality, accessibility and correctness of the information (MSB 2012). Within municipal information security, “documentation in social care” is seen as one of the ways to access information that the security work should take into account (ibid.)
1 See the Act (2006:544) on the responsibility of municipalities and county councils before and
during extraordinary events in peacetime and in times of heightened preparedness, and the accompanying regulation (2006:637).
2 Here I use the broad concept that is used within the social services and within health care (see the
terminology database of the National Board of Health and Welfare) and that encompasses what the law (see note 1) calls “extraordinary events.” I do this in order to emphasise that the municipal crisis preparedness is not only concerned with events entailing important consequences but also with events that, from a societal perspective, could be described as having a lower impact (cf Cuadra 2016).
3 Crisis preparedness is defined in regulations (MSBFS 2015:5) and refers to the ability to prevent,
resist and handle crisis situations through measures and a structure created before, during and after a crisis.
4
Regarding research in this area, there are Swedish studies with a combined interest in documentation and IT within the social services, but these studies concern everyday situations, that is, not failures. Thus, one dissertation deals with the implementation of systematic documentation for the development of the operations within the social services (Alexanderson 2006). Furthermore, the effects of the computerisation of elderly care have been studied from a computer science perspective (Hedström 2004). Studies that can be found in the
international literature4 regarding documentation within social work also give the impression of focusing on the everyday performance (e.g., Reamer 2005;
Cumming et al. 2007; Hansen et al. 2015). There are, furthermore, studies of the use of IT that goes beyond documentation (Csiernik et al. 2008; Reamer 2013; Fitch 2015; De Rosa 2017) and of the effect that IT has on social work
(Gillingham 2016; Perron et al. 2010).
As for studies of failures in information technologies, I cannot find any that take an interest in social work (e.g., Abbas & Pinsonneauelt 2016; Cohen 2000). I do, however, find studies that are interested in the same type of failures within health care, in terms of patient safety (e.g., Magrabi et al. 2016), which has a certain bearing on home care with regard to medical records. Finally, I want to mention that documentation was touched on in a Swedish study of the social services and crisis preparedness (Cuadra 2017b), but merely based on the fact that the legislation of crisis preparedness requires that measures be reported,
something which presupposes documentation.
Against this background, the aim of my investigation is to contribute to knowledge about what IT failures can mean within the social services. The focus is on the immediate consequences for the documentation and communication of the social services, as they have been formulated by social services personnel in risk and vulnerability analyses performed within the framework of the
investigation. The aim is also to contribute to the understanding of what happens during failures. The function of social work is used as a starting-point with a
4 My assessment is based on searches in the databases Scopus, Sociological abstract and
Libsearch. The search words were “information technology,” “internet,” “failure,” “outage,” “interruption,” “breakdown,” “documentation,” “social work,” and “social services” in different combinations.
5
special focus on the relation between the events and the maintenance of the function despite interruptions.
Next, the theoretical approach that has contributed heuristic tools, in the shape of perspectives and concepts, will be presented. Thereafter, methods,
material and procedure will be set forth, followed by the result of the investigation and my interpretation of the results. I will end by drawing a number of
conclusions, taking into account the limitations of the investigation.
A systems-theoretical approach
I have chosen a systems-theoretical approach. This means that social work is seen as a specific social system (Michailakis & Schirmer 2017) which enables a conceptualisation of the events that I try to understand. It gives an insight into the social services, partly in terms of systems levels, partly in relation to their
environment, and finally as a body “aimed at solving specific problems” (Michailakis & Schirmer 2017:75) and as part of a functional system.
Systems theory is communcation theory (Jönhill 2012). Communication is a process characterised by the selection, transmission and understanding of
information in a certain way (Appel Nissen & Magnussen 2013). Communication is never completed; it gives rise to new communication in a perpetual circular movement (Jönhill 2012) through equally perpetual feedback that changes the system and its operations (Michailakis & Schirmer 2017:34). It is communication that keeps social systems together (Jönhill 1997). It also maintains the systems’ connections to the environment from which the system at the same time attempts to differentiate itself (ibid.). The differentiation takes place through the
maintenance of those borders that preserve the system’s own identity and existence (Maturana & Varela 1980). The environment is in that sense the negative correlate of “everything else” that the system can be in contact with (Luhmann 1995:181). By viewing systems like this, that is, as operationally closed in self-referential processes that simultaneously provide an openness to the environment from which the system tries to differentiate itself (Luhmann 1995), the importance of their continuous communication with the environment is brought to the fore. The environment could, as in the case of this investigation, be that of the social services which is peopled by persons for whom the social
6
within other social systems, and possible examples here are the Administrative Court and the Swedish Public Employment Service.
Social work as a functional system
The course of events that the investigation touches on may, using Luhmann’s thinking as a starting point, be seen as taking place on different levels of social systems; interaction systems, organisational systems and the societal system. The levels could be compared to Russian Babushka dolls. In my version, the innermost doll contains the social services staff that interact and talk in mutual presence (or over the telephone) with their clients, users and colleagues. In connection to the conversations and the documentation of those conversations, the staff use so-called distribution media (Michailakis & Schirmer 2017: 60), such as printed texts, telephony, computers and the internet. Distribution media allow for a disconnection between text and author but they also make it possible for data and information to be preserved and distributed (ibid.). I will return to this theme.
The next “doll” is constituted by the social services as an organisational system with their “arrangements for decisions and decision premises”
(Michailakis & Schirmer 2017: 53). They can impose certain rules, themes and prestructurations on the interaction systems, even if the interaction systems are at the same time autonomous to a certain degree (ibid.). Distribution media are resorted to, for example, when a social service worker in the context of the organisational system sends a fax to the Administrative Court or when a head of unit in the elderly care “approves” a decision about social support for
implementation. With regard to the investigation, it is important to note that each organisational system may in its turn be enshrined in other organisational systems (Jönhill 2012). The social services are enshrined in the larger municipal
organisation and may at the same time be broken down into their internal operational areas, departments and units (see Michailakis & Schirmer 2017).
Finally, the outermost “doll” is the modern societal system. Note that it is seen as differentiated into several functional systems (Luhmann 1995). These are autonomous in terms of communication, decisions and expectations (Appel Nissen & Magnussen 2014:127). It should also be noted that Luhmann never discussed social work in terms of a functional system of its own. However, leaning on newer work, I choose to see social work as a functional system (Appel
7
Nissen & Magnussen 2014; Schirmer & Michailakis 2015; Michailakis &
Schirmer 2017). 5 Social work is also called organised social aid and seen as part
of society’s general aid system together with health care (Roth & Schütz 2015). As the term implies, a functional system is characterised by its having a function in the sense of “being a body aimed at solving specific problems”
(Michailakis & Schirmer 2017: 75). In the case of social work, that function is the
administration of exclusion (Michailakis & Schirmer 2017) or the management of
exclusion (Schirmer & Michailakis 2015), given that inclusion is what constitutes the link between people and society (ibid.). Through striving for the vicarious
inclusion of persons, the function of social work is to counteract the destructive
effects of exclusion from other functional systems (Jönhill 2012), effects that may have cumulative consequences. Inclusion means having access to what a
functional system provides (ibid.). Social work can, to use Jönhill’s example, address the exclusion from the economic functional system that unemployment constitutes, or the exclusion from the care and love of the family (as in the case of child abuse) that I myself connect to Luhmann’s examples of functional systems. Just like politics, the legal system, science and education, the family is part of those systems (see Luhmann 1995).
Functional systems, just like social work, are characterised by having their own communication medium and a binary codification. The code in question reduces the complexity of the world and of human beings to a “small window of relevance” (Schirmer & Michailakis 2015). The respective medium of all the functional systems relates to what the Other may in some cases experience (e.g., family love, or the value of truth in science) and to what the Other may, in other cases, act upon (e.g., money, or her/his right) (Jönhill 1997). Where the economic functional system uses money as a communication medium and payment/non-payment as a binary code (Jönhill 1997), the medium of social work has been suggested to be help (Appel Nissen & Magnussen 2013). According to my interpretation, on exercising this help, social work resorts to the medium of the other functional systems; for example, a social service worker writes a statement
5 There is an ongoing discussion of whether you can complement the functional systems that
Luhmann wrote about (personal mail exchange with Jan-Inge Jönhill). The discussion also concerns how to conceptualise the functions, codes and medium of social work (this will be developed below). I have the impression that this discussion has been conducted for several decades and largely within German-speaking literature.
8
to the rent tribunal in support of the Other’s right to their lease, and the social services pay.
Regarding the binary code of social work, need/non-need has been suggested (Appel Nissen & Magnussen 2013). I choose to specify the need referred to as concerning persons’ social address in relation to other current societal functional systems. The economic functional system can once again serve as an example; through special measures, social work helps the Other, an unemployed person, to obtain a social address and thus to be seen as communicatively relevant or, more exactly put, as a relevant addressee for communication (Schirmer & Michailakis 2015) in the economic system. The function of social work can thus be specified as a “re-set of social addresses” (Roth & Schütz 2015: 24). This could be done through an internship or through payment of money in the shape of income support.
Systems can develop structural couplings among themselves (Maturana & Varela 1992). In Luhmann’s version, such couplings can be developed between functional systems. They do not concern material structures but rather connections between selected themes for communication (Jönhill 1997). For example, the functional system constituted by the legal system may discuss the theme of family relations even if it is not in itself the functional system ‘family.’ Viewed in this way, the communication of the social services with, for example, the police, the Administrative Court, school and the Swedish National Insurance Agency, is just that, structural couplings.
Trivial machines, information and documentation
Above, I referred to printed text, telephony, computers and the internet used within the social services on handling documentation and communication, as distribution media (Michailakis & Schirmer 2017). In order to indicate the difference between this kind of distribution media and alternative technological systems approaches that use the concept of system when referring to the
technology that enables such media, I opt to call the latter trivial machines (Michailakis & Schirmer 2017: 32). In Luhmann’s sense, the term system is reserved for social systems.
So how should the concept of information be understood in the investigation regarding failures in information technologies? An initial answer is that
9
information is “a selection from a (known or unknown) repertoire of possibilities” (Luhmann 1995:140). Information reduces complexity by expressing a selection and thereby excluding other possibilities (ibid.). From this perspective, I see documentation as information that is “packaged” in a certain medium and that entails a certain security for the staff. The trivial machines come across as an “uncertainty-reducing mechanism” by packaging the “expressed selection” of information. I imagine that when the technology fails, the number of possible alternatives increases and thereby also the uncertainty. Maybe a social worker who cannot access the record wonders, “Was it in June or in July that the placement of the child took place or was it at an earlier date?” In this context, perhaps a sticky note with a hand-written comment may have a “functional equivalence to absorb uncertainty” (Luhmann 1995:185). The example is chosen in order to highlight the concept of functional equivalence. Being able to maintain a function through such equivalence is in several ways to do with redundancy,6 that is to say, lost units may, thanks to a multitude of units, be replaced by others (cf Luhmann 1995). Perhaps documentation can take place by means of pen and paper if the trivial machine fails?
Method
The investigation is based on risk and vulnerability analyses carried out within the social services in four municipalities in the spring of 2015. The municipalities participated in a development project under my guidance, and the aim was to strengthen the role of the social services in the societal crisis preparedness. In the working groups formed in each municipality the participants were employees within all the operational areas of the social services, that is, individual and family care, the elderly and disability area, including social psychiatry, and the municipal home care. Among the members of the working group, the number varying
between five and nine persons, there were heads of unit, quality developers, housing supporters and medically responsible nurses.
Within the framework of the developing project, five workshops were carried out, three of which were organised as risk and vulnerability analyses. This
6 Redundancy can, using Luhmann’s words, both “guard against losing resources,” as in the case
of spare power units, and, as in the case of system internal bases, contribute to security through, for example, internal guidelines and templates that “might be suitable” (Luhmann 1995: 184).
10
investigation is based on the group conversations taking place during one of those analyses in each municipality. The theme of the analysis was a longer period of IT failure.
Risk and vulnerability analyses are an analytical method used on a regular basis within the field of crisis preparedness (MSB 2011). They are done based on the legal requirements mentioned in the beginning and according to different methodologies. The method can make visible the risks of an activity (the possibility of negative events), its vulnerabilities (the susceptibility to adverse effects) and its ability to handle the outlined situation in all of its complexity. With regard to vulnerabilities, they are closely related to dependencies (e.g., on certain resources, on colleagues or on the management), which may thus also come to the fore in the analysis.
In my case, the starting point was an already identified risk in the shape of a fictive future negative event, that is, a scenario. The scenario was written as a gradual course of events and produced by a council administrative board with a view to being analysed in municipalities.
On each occasion, the respective groups gathered in the workplace during approximately three hours. The participants were given the scenario in writing and we went over it, step by step, under my guidance. As a guideline, we had the following questions:
How and when is the event noticed in the activity? Who notices it? What in the activity is difficult to maintain?
What needs to be done? By whom? What questions and uncertainties arise?
Who are our most important cooperation partners at this stage? Take into account both those within and those outside of the activity. Who may need information from and cooperation with us?
What is the activity dependent on? That is, what is needed in order to be able to … ? Why? It can be, for example, certain resources, agencies, knowledge.
How critical is this dependence? Are there any “shock absorbers” that also provide endurance?
What vulnerabilities can we identify? Where within the activity can we discern vulnerabilities? Describe the vulnerability in as much detail as possible.
Have we got suggestions for measures that may reduce the vulnerabilities that we have identified?
What is the expected effect of the measure/s?
Who can implement the measure? What would be required in order to make it happen? The conversations were recorded, besides each and every participant, including me, being expected to take notes. The recordings were transcribed in their entirety. This documentation then constituted the basis for a thematic summary
11
(see Spencer et al. 2014) and for my initial processing of the material, where I focused on the consequences of the scenario for the documentation and
communication of the social services. Parallel to this, I developed my theoretical perspective, by means of which I could conceptualise observations and also deepen the analysis of the course of events during failures.
The initial summary was discussed with the participants in each working group in order to ensure an agreement with the assessment of the participants. The summary was given the following structure: what happens; what work tasks are concerned, what actors take action; vulnerabilities and dependencies, and, finally, suggested measures.
The scenario was mainly about a municipal network that was inaccessible for two to six weeks. It is reproduced below in a somewhat summarised form:
Stage 1. The initial phase: One Tuesday morning the IT unit is notified by a user that there is a
problem connecting to the network of the organisation. Calls about similar problems pour in. The most common problem is that the user accounts do not work and that the network is slow. The spread within the organisation is rapid.
Stage 2. The troubleshooting phase: The reason for the problem is unclear. It could be the result of
a denial-of-service attack or a virus. The IT unit decides to close off the network that very evening. It becomes impossible to access the network. The time span is unknown.
After several days the problems have been identified as caused by a virus. This has probably happened through a memory stick. It is still unclear what damage the virus has caused in the network, but since all the computers in the network are linked to each other through the
authorisation to access the operational systems and file servers it is difficult to isolate the effects.
Stage 3. The endurance phase: The IT unit works for a week trying to reset the network.
Representatives of the antivirus programme and security experts arrive and work together with the organisation to fix the problems and minimise the effects. The assessments of the time needed for the work vary, but another two to six weeks will probably be required. During the resetting, the IT unit discovers that numerous files that are of importance for several different departments have disappeared and that the content of other files has been altered. As a result, there is considerable uncertainty regarding whether the information in the internal systems is reliable.
The scenario gives an undisguised impression of being written by experts at information technology, which in itself constitutes an important quality. I also had the opportunity to talk to IT managers before the analysis of the scenario, and they shared their perspectives on the meaning and handling of the scenario with me. I
12
now proceed to the presentation of what the risk and vulnerability analyses indicated.
Results
Since the investigation does not have any ambition to compare the participating municipalities, the results are presented together. This is facilitated by the fact that the assessments made by the four working groups had considerable similarities, even if there were certain differences, for example, with regard to the number of operational programmes affected. I want to highlight the fact that in some of the working groups there were persons who had experiences of similar though less serious IT outages. There was also the experience of, as a head of unit expressed it, “stage one [in the scenario] not being unusual […]. It happens once every month or so.” Such experiences constituted useful references during the group conversations.
I will start by giving an overview of what happens, chronologically, based on my summary. Even if documenting practices are touched on in this introductory section, I have chosen to develop the consequences for documentation in a separate section.
I will, by means of the systems-theoretical perspective and its conceptual apparatus, show how the technical failure leads to what I call adaptations. They are partly about replacing lost units based on functional equivalence and
accessible redundancy, and partly about switching between systems levels and about making prioritisations, that is, new selections of possibilities. I will point to switches between the “Russian Babushka dolls” in terms of interaction systems (in mutual presence) and organisational systems. Within the latter level I will differentiate between the intraorganisational level of the social services and the municipal intermediate organisational level, which encompasses the social services. The relation to other functional systems surrounding the social services, and that the social services communicate with, will also be highlighted. I will assert that, through making adaptations the social services can maintain their function, though not wholly without implications for the staff as well as for clients and users.
13
Overview of the course of events
Stage 1. The initial phase: In the initial phase, the failure would make itself felt
through difficulties logging into operational systems. As one participant expressed it: “The user account is the first problem and then everything is a problem.” Since the social services conduct their operations around the clock, the IT problems that the scenario depicts could occur outside office hours as well. In such cases no error report could be communicated. This was seen as a vulnerability of the operations.
In this phase, the participants perceived a dependence on the IT unit. The dependence concerned not only technical support but also adequate information in order to avoid rumours and in order to be able to maintain a realistic attitude to the situation and take into account a reasonable time frame. The management of the social services was expected to keep up the intermediate organisational
communication with the IT unit. Furthermore, the task of the management was considered to include the responsibility for the communication with the
environment of the social services with regard to family members and users. Note that the nature of the physical problem was such that it was impossible to use the computers at all, even as “typewriters.” This meant that the entire staff must write by hand. It would not be possible to print or to use what is commonly called digital faxes either.
Stage 2. The troubleshooting phase: When the network was closed down,
the operational systems became inaccessible. This concerned, for example, the record-keeping system and systems for intervention planning and documentation, coordinated record keeping, coordinated care planning, coordinated individual plans, deviation reporting, medical prescriptions, the palliative register and the programming of safety alarms7 as well as those programmes that occupational therapists use for ordering aids. In certain cases, such as coordinated care planning, the programme function was considered replaceable by what I call shifting down with regard to complexity and instead switching to the interaction system and to a telephone call. This also applied to the programmes that were used by the municipal large-scale catering establishments for ordering goods, as
7 Existing safety alarms were not considered to be affected by the failure, as these were not
dependent on the municipal network. I want to point out that this state of affairs does not exclude vulnerability in relation to other networks and alternative scenarios.
14
well as the programmes used by the different activities for ordering food from those establishments.
The same switch, from the organisational to the interactional level, was expressed by a head of unit within elderly care who discusses how co-workers document door codes. She says that “my housing supporters enter door codes in
[named record-keeping system] […]. But I think it’s a minor problem, because I think that the codes we have, they are in the heads of the staff. You can probably manage the most acute phase; if you can’t manage on your own, you phone your colleague who was there last week. It’ll be much more time consuming but not too difficult to handle.” However, such a switch between systems levels was not available with regard to the programming of new safety alarms, where no possible measures or alternative units could be identified.
In the second stage of this scenario, the working groups also talked about being dependent on the prioritisations made by the management. The success of such prioritisations was seen as having consequences for the rest of the sequence of events. On an intraorganisational level, and in concrete terms, this meant how the documentation should be done, and within my theoretical framework it meant making new selections of information in order to avoid arbitrariness. The
dependence was also due to the fact that the management in an intermediate organisational group concerned with the whole municipality, was expected to make decisions regarding the order of priority in which the computers would be reset (that is, cleaning every computer, one at a time, from viruses by reinstalling all the programmes). In some municipalities, the order of priority also concerned access to the “clean” computers that could be made available by the IT unit during the failure, using an alternative network. This could perhaps be arranged in a schoolroom where a network has been made available. I want to mention that the working groups raised questions about how well this solution, unilaterally planned by the IT unit, could work, since it was assumed that the social services staff would move between their workplace and the make-shift computer room. The working groups’ questions also concerned who would “have first dibs” and what the criteria for such a prioritisation would be. One suggested criterion was to prioritise the staff whose work was to collect and coordinate data from different programmes, for example, a schedule-planning programme that combines data about users and staff. That is to say, the suggestion concerned employees whose
15
documenting practices were particularly complex and had an intraorganisational communicative function – practices for which there were deemed to be no available alternative units with a functional equivalence.
A number of programmes used by the management for administrative purposes would also become inaccessible during this scenario. Examples of such programmes are salary systems, invoicing systems, budget systems, occupational injury reporting, internal registration systems and systems for staff-related matters, and programmes for the staffing and for the front office’s messages to case workers. Several of these programmes are also used for intraorganisational communication through shared authorisations. No alternative solutions other than returning to manual management were identified. The participants in the working groups did not think that work tasks performed by means of this type of
programme should be prioritised, maybe because those work tasks did not directly concern the function of the social services, that is, work with the social addresses of clients and users.
Another consequence of the operational disruption was due to the programme Outlook being used as a calendar. As a result, the planning already done became inaccessible. The working groups realised that there was a risk of appointments and home visits being missed, even if it was likely that they would be contacted by users and be reminded by them. The vulnerability that Outlook constituted was thought to be remedied through all the employees systematically and regularly printing their planning and/or complementening it with traditional calendars. The latter solution was already resorted to. One head of unit within the individual and family care illustrated the present situation when saying that “my staff do it both ways, they don’t trust the system.” Furthermore, it was believed that the vulnerability could be remedied through “reconstruction from memory” and through the staff having a good knowledge of the persons concerned. That is to say, the suggested measures contained adaptations that entailed a switch to the interactional systems (such as telephone conversations with clients) that allow for a functional equivalence, albeit less elegant. I note that the intraorganisational function offered by Outlook (that is, the possibility to see one’s colleague’s calendar) was not prioritised. This means that maintaining the social work with clients and users was prioritised.
16
Yet another consequence of the technical failure concerned voice over IP. Questions were raised about whether it was possible to use the mobile extension function (in order to be able to use the telephone as an “ordinary mobile phone”) and if so, if the staff were sufficiently familiar with it. The issue was expected to be addressed intraorganisationally by the management and be handled collectively and thereby interorganisationally, that is, municipally, through the switchboard of the municipality.8 One of the municipalities had already made an adaptation with respect to voice over IP. The municipality in question had strategically chosen to have a mobile phone with a contract for each care facility as part of their business continuity planning,9 precisely because of the vulnerability of voice over IP. However, the working group of that particular municipality concluded that an inventory needed to be made of who currently had those telephones, as well as an investigation of whether the telephones in question needed complementing (for example, with Rakel mobiles).10
A related question was whether the collective text messages that are used to contact substitutes would work and whether the contact information to the substitutes would be accessible in case the failure was such that there was no opportunity to use mobile extensions.
Digital faxes are also affected by IT failures. They are used regularly in the communication with organisations within other functional systems, as well as within one’s own system. A quality developer described the importance of this communication for maintaining the function of the social services like this: “if the fax doesn’t work we can’t receive notifications of concern, reach the social emergency office, the police and criminal justice . . . those are heavy issues.” She emphasised the seriousness of the matter by adding that “we had a fax that was out of order during one day and that was difficult enough, after all we don’t know what agency out there needs to be in touch with us.” She complemented her
8 In the analyses, this was assumed to be not only possible but also carried out, which is why
telephoning was assumed to work.
9 Business continuity planning aims at ensuring the ability to “deliver” at what is considered a
“tolerable level” regardless of interruptions. It means making prioritisations among one’s undertakings and producing plans for how to maintain the operations or at least reduce the duration of the interruption.
10 Rakel is a communication system that has been developed in order to deal with crisis situations
and interruptions and that covers the whole of Sweden, and it is considered to have a high level of operational security.
17
enumeration of surrounding organisations by mentioning the Swedish Insurance Agency and the Swedish Public Employment Agency.11
The assessments of the consequences that the technical failure would have for e-mail varied between the municipalities, depending on their technical
solutions. However, e-mail was not given much space in the analyses, since there was a measure of uncertainty regarding the actual circumstances.
Stage 3. The endurance phase: In this stage (week two to six), onerous manual
documentation routines were deemed to have consequences for the working environment. The input value was that the staff are “already, even under normal circumstances, behind with the documentation.” The situation was seen as the potential “last straw” for an already burdened staff and as potentially leading to (an increase in) sick leaves.“We mustn’t forget the reality we live in, the staff are already streamlined. They have been slimmed down from 36 to 26 staff members in my department,” a head of unit pointed out. However, the seriousness of the consequences was directly linked to the success of the management’s
intraorganisational adaptations to the situation, in the shape of prioritisations, staff management and other decisions.
Stage 4. The stage added by the social services: Interestingly enough, the
participants in the working groups added a stage to the scenario. It concerned the considerable amount of work that needed to be done after the network had been reset, work consisting of entering the documentation that had been done manually in digital records. This stage was called “resetting,” using a concept borrowed from stage 3 of the scenario. Even if this step may seem to be a purely technical aspect of the scenario, the working groups clearly prioritised it. This will be developed below.
Consequences for documentation
For the documentation of the operations, the immediate consequence of the closed-off network and of failing trivial machines is that record keeping becomes impossible. Handling the record-keeping system via the telephone would not be possible either, so the immediate adaptation in order to reach functional
11 My preliminary understanding of these organisations is that they are to be found within the
18
equivalence would, as mentioned earlier, be to switch to writing by hand. In order to give the notetaking the right dimensions – such as resorting to a ring notepad, instead of using separate sheets of paper in the belief that the problem will be very temporary – the staff rely on information from the IT experts via their
management. The assessment made by the working groups was that without adequate information the co-workers would think that “everything will be back to normal tomorrow…” and that it would therefore take some time before the documentation would become systematised. Above all, prioritisations were seen as having to deal with what and how much to document by hand.
In order to facilitate the transition to a manual handling of the documentation, it was suggested that templates for different purposes should be produced within the framework of the crisis preparedness – primarily in connection with decisions regarding, for example, immediate placements of children and social welfare, but also regarding other kinds of record keeping and document registration. Some participants, however, believed that templates are unnecessary. “You can pick a sheet of paper with a logo on it, we do after all know what it’s supposed to contain [ . . . ] then it will be fine.” The question of routines for the correct safekeepingof the handwritten documentation was raised, and such routines were deemed to be already in place, not least through the availability of filing cabinets and archives. A more long-term measure that was suggested was the provision of a small number of “stand-by computers” (“stand-alone,” that is, without internet connection) with local printers available for prioritised printing.
In a similar scenario, the joint use of a record-keeping system for
documentation would, as mentioned above, be impossible. This would affect, for example, operations within elderly and disability care, such as home care, housing support and evening/night nurses who are not collocated and who communicate via the record keeping around users with several different interventions. Thus, a special task would be to share information despite an IT failure and without losing anyone in the process. The seriousness of the situation was expressed like this by a nurse: “users [ . . . ] risk being wrongly medicated and misjudged and things could turn out badly.” The suggested adaptation meant switching over to the interaction system, that is, to telephone conversations. The same problem had arisen in relation to the social emergency service in one of the municipalities. This
19
means that the communication with the social emergency service would also have to switch to the interaction system and take place via the telephone.
Due to inaccessible record systems the staff would not be able to access the history of the records, which was seen as a potential cause of serious
consequences in individual cases. Within individual and family care, for example, it was believed that the handling of cases would risk losing valuable time with regard to both child care cases and cases of substance abuse. It could also lead to delayed reports to, for instance, the Health and Social Care Inspectorate and the Administrative Court (and thus to a request for extended time), since the
investigations that the reports rely on were not printed. Immediate placements were considered possible in very uncertain cases, cases where the access to the record would have contributed decisive information. Here, once again, I see a switch to the interaction system, that is, to personal contact with the client and the agency or court, respectively.
Within home care nursing, the overall challenge was, according to a
medically responsible nurse, “to uphold the activity without endangering the legal and patient safety.” In home care nursing, an inaccessible record could result in hospitalisation. That would above all apply to on-call time and the encounter with “new” patients that the nurse did not know from before. With regard to the
problems connected to inaccessible record history, it emerged that there was an existing business continuity planning that consisted in “important events” being printed as a routine procedure. Concerning everyday routines, the staff had thus tried to make themselves less dependent on uninterruptedly functioning “trivial machines” in order to access text. “We have routines [. . .] that say that everything should be printed, important events, in the persons’ records, but this is often old information, it makes for a duplication of our work,” a nurse explained. Printing was thus not seen as a sufficient measure that could completely eliminate the vulnerability. It was considered difficult to sort out what is important; data soon becomes obsolete, so the vulnerability was deemed to partly persist. The same procedure, that is, printing, applied to the nurses’ schedules and the lists of whom they are supposed to visit daily.
Despite the negative consequences of inaccessible records, the home care nursing was considered less vulnerable in the current scenario. That assessment
20
was based on the fact that medication lists are stored outside the municipal networks and exist in a national web-based prescription database.
Other circumstances within home care nursing that contributed to less dependence, relatively speaking, on trivial machines, were, for example,
handwritten referrals being sent with the patient in case of an ambulance ride and complemented with conversations both with the ambulance staff and (on the telephone) with the hospital in question. Furthermore, patient care planning was largely done without the support of trivial machines, and care plans (between the municipality and the health care, on the patient going home from hospital) were considered possible to establish over the telephone.
Neither were the LSS (the Swedish Act concerning Support and Service to Persons with Certain Functional Disabilities) housing and housing support seen as particularly dependent on trivial machines for accessing already completed
documentation. Here too, printed implementation plans (spanning approximately six months), weekly schedules and medication lists made the situation
manageable.
Similarly, the elderly care was not considered particularly dependent on technology, even if the vulnerability varied between the municipalities in case of inaccessible record history. The care administrators usually wrote by hand and subsequently registered the decisions in the record, and they normally also communicated decisions to users over the telephone. A head of unit within the
elderly care says that “we have a routine for my group according to which we
must record all persons in a locked-up folder, those who receive support, they should be updated. We have names and personal identification numbers, the person’s card,” adding that “the personal files [. . .] are updated every day. We keep all the personal files in the archive, we print all decisions and put them in the file.” This is yet another example of how printed data is used routinely as part of business continuity planning in order to reduce the dependence on IT. The dependence on the proper functioning of trivial machines has thus already been reduced in the everyday routine activities.
Then, as regards the “resetting” (stage four), the assessment differed between the municipalities concerning the extent to which it would be acceptable to scan handwritten text or whether it too would, just like decisions, have to be digitally registered in every record in question. The extent of the process was considered
21
directly linked to how prioritisations had been made during the initial phases of the scenario. Nevertheless, this task was seen as considerable and burdensome and as something that could last for several weeks, and a head of unit pointed out att
“if we had asked the case workers to scan, cut and paste, etcetera, they would have gone mad. We must consider who is supposed to do what. It would be costly in terms of working hours and well-being, and a burden.”
In this final phase, a clear interorganisational dependence on the IT unit was identified. This phase meant, according to how the scenario was written, that the staff were expected to be able to determine whether information had disappeared. The working group expressed uncertainty regarding whether they themselves or their colleagues would really have been able to do this in all types of programmes.
A head of unit within the individual and family care believed that “if all the care
recipients in the town centre were located in the western part, this would be discovered. But if it’s a matter of deleted notes, I don’t know if it will be [discovered].”
At this stage, questions arose that were expressed in terms of “the possibility of users disappearing from the digital box” before they are “approved,” that is, in the programmes used by decision makers and where the executive staff confirm their receipt. One head of unit stated that this inter- and intraorganisational joint use made it “important not to move a person [from hospital to housing] until this has been confirmed orally by the hospital and the short-term accommodation.” She went on to state that “it can be disastrous if you have something in the digital box that has to be implemented within ten days and then that file disappears . . . then when it’s time to implement those interventions no one knows about them.” The suggested measure was to always complement the digital communication with oral contact. “That’s why it’s so important that this interpersonal contact between the hospital and the home care service exists. On handing someone over there has to be oral contact,” said a head of unit in the home care service who verbalised the switch to an interactional system “just at the interface between an organisation and its environment” (Michailakis & Schirmer 2017:58).
Income support
As the documentation that is linked to social income support is dependent on the network, IT failures will have consequences for the possibility to make payments.
22
This concerns emergency income support and payments to foster families as well. The gravity of the scenario was emphasised by a quality coordinator in the
following terms: “if this occurs at the end of the month when everyone is waiting for support, there may be a crisis. We don’t have any cash. If we’re going to write checks we must access the operative system. The same goes fortravel cards and supermarket cards.”
With respect to income support, decisions about support and about payment are written electronically and files are sent to the economy department, also electronically. The adaptation in this situation meant switching to a handling that is independent of trivial machines and, thus, writing the decisions on paper. The adaptation also meant switching to the interaction system “at the interface
between an organisation and its environment” (Michailakis & Schirmer 2017:58) through a case worker making phone calls to landlords about delayed payments. Handwritten requisitions also occurred as an alternative form of payment, while other municipalities resorted to giving the individual person a pre-loaded cash card (for use in cashpoints or in a certain chain of food stores). The latter technical solution led to a suggested measure, namely, having spare pre-loaded cash cards “in stock.”
The fact that the records would be inaccessible would entail that case workers would have to “trust what they [the clients] say.” A head of unit remarked that “you must be prepared to pay too much when something goes wrong.” Therefore the resetting phase was also judged to entail adjustments of erroneously paid income support.
The maintenance of the functional system
A number of adaptations have been described, as well as measures suggested regarding similar situations in the future. The gravity of the scenario was
emphasised in the working groups with reference to the fact that it had given rise to a Lex Sarah report on risks of maladministration, in this case due to a
technological failure (the National Board of Health and Welfare 2014).12
12 One example from the handbook of the National Board of Health and Welfare regarding what should be reported concerns the breakdown of digital documentation systems. In the example it says that in such a situation the employees will not be able to access current documentation according to SoL (the Social Services Act) or LSS (the Swedish Act concerning Support and
23
The adaptations are to do with the replacement of lost units, but also with downshifting with regard to complexity and switching to interaction systems, as well as making prioritisations. In my view, all of the adaptations may be
understood as contributing to the maintenance of the contacts of the social services with the surrounding society and the functionality of the social work, by enabling continued work with the social addresses of users and clients. It emerged that there existed a degree of business continuity planning that also reduced the dependence on trivial machines. It consisted of systematic printing and strategic “ordinary mobile phones,” oral contacts and an analogous fax machine or two but also of knowledge of people and having “door codes in one’s head.”
The working groups did – as is customary in a risk and vulnerability analysis – an assessment of the consequences for the activity and of its ability to handle the scenario. That ability varied between being overall good – at least in practice, thanks to strong individual resources and constructive and practical solutions – and deficient or even very deficient. According to the working groups’
assessments, the scenario may have serious conseqences for individuals’ life and health, depending on how the situation is handled (for example, remedying wrong medications on the telephone, or hospitalisations and immediate placements being implemented to be on the safe side), even if the assessments differ somewhat between the municipalities and the different operational areas.“My assessment is that it won’t be hard for the patients but that it will entail an incredible amount of work for managers and others. But those who meet patients will deliver what they are supposed to deliver,” said a medically responsible nurse in the home care nursing, thereby making a good summary of the total assessment. The activity can
“deliver,” although at the cost of some extra work and delays. However, the staff’s trust in the operations of the activity is considered to be at risk of being affected negatively if the management is not able to prioritise and lead the activity through the scenario as well as communicating with the IT unit. The trust in the
Service for Persons with Certain Functional Impairments), which may result in interventions not being performed at all or not performed in the correct way. Information from the day staff about a person having come home from hospital and being once again in need of home care will not reach the night staff, and employees who handle economic support will not be able to deal with new cases or pay out the support that has been granted to the persons entitled to economic support (National Board of Health and Welfare 2014).
24
area of elderly care and disability was judged to be less affected than the trust in individual and family care, since, as a quality developer put it, “people’s trust in us is already low [ . . . ] and if we can’t act practically without a computer their trust will decline even more.” The differences for the different activities with regard to the consequences of the scenario may be due to the fact that they, in their work with clients’ and users’ social address, communicate with different societal functional systems. While those that deal with income support
communicate with the economic systen, those that deal with care communicate with the family,in Luhmann’s sense. As a head of unit in elderly care expressed it: “We make interventions as persons for another person. You [income support] make an intervention for the person in relation to the system, [ . . . ]. That is, they want something, right, it isn’t always help from a person, they want something that the system gives them, like money.” I would argue that, during the scenario in question, the work with the social address of clients and users was to a certain extent affected in its prerequisites, even if the social work was maintained as a functional system. This manifested itself in, for instance, the risk of visits being missed, the risk of delayed investigations, delayed installations of safety alarms, erroneous medications, and hospitalisations and immediate placements in foster care as a precautionary measure. Other manifestations of the effect on the social work were the use of potentially stigmatising requisitions and delayed rent
payments, as well as a possible decline in trust in the activity and its management.
Conclusions
This investigation would have been strengthened by also dealing with the perspective of the management and by having further access to the knowledge possessed by the IT managers. Nevertheless, I believe that the investigation has shown that IT failures entail huge challenges for social work as a functional system. My conclusion is, however, that the IT failure in the scenario in question does not lead to a corresponding failure in the function of social work, despite the technical dependencies of the social services. The function is maintained through adaptations, which are possible thanks to functional equivalence and available redundancy. I want to draw particular attention to the fact that it is the staff that, besides technical adaptations, make concrete interactive adaptations, to some
25
extent at the expense of their working environment. I call this social redundancy13
in order to highlight how lost technical units or trivial machines can be replaced by social interaction. My conclusion is that social redundancy seems to be what can replace lost technology.
With respect to adaptations, someone said that “keeping the paper system in case IT collapses isn’t a natural way of doing things, it’s not sustainable,”
referring to the amount of printing. Despite this and despite ironic comments on “the paperless society,” printing appeared to be one of the adaptations at the disposal of the staff. The way I interpret this, adaptations are limited to what you “own,” are knowledgeable about and can influence.This impression is
strengthened by comments such as “the IT people don’t have a clue what we use our systems for” and the view that the choice of technology is not always based on the requirements of the activity. Those perceptions support the conclusion that there seems to be a space for developing the interorganisational relations with the IT units, whose technological know-how determines the choice of technology.
Information security is increasingly relevant (MSB 2015; see also MSB 2017). In view of the central place of information technology in the
documentation of the social services and other documenting practices, information security seems to be a development area within the social services. A possible avenue to explore in this respect would be to develop structural couplings (Luhmann 1995) between the operations of the social services and the functional system within which societal security is handled, by meeting in a joint theme related to the function of social work, namely, dealing with social exclusion. This could perhaps be initiated through the participation of the social services in municipal risk and vulnerability analyses within the field of information security. Thereby the social services would be able to share their perspectives on the events during IT failures and their consequences and have an influence on both risk assessments and technical solutions, as well as further strengthening their ability to handle such failures without serious consequences for their operations.
13 The concept derives from conversations with Jörgen Sparf, Risk and Crisis Research Centre,
26
Literature
Alexandersson, K. (2006). Vilja, kunna, förstå: om implementering av systematisk
dokumentation för verksamhetsutveckling i socialtjänsten [Willingness,
comprehension, capability: about implementation of systematic documentation for developing social work in the public social-work services]. Dissertation, Örebro University, School of Law, Psychology and Social Work.
Appel Nissen, M., & Magnussen, J. (2013). Niklas Luhmann – en kompleks teori om et komplekst samfund [Niklas Luhmann – a complex theory about a complex society]. In S. Juul Hansen (Ed.). Sociologi i socialrådgivning og socialt arbejde [Sociology in social counselling and social work]. Copenhagen: Hans Reitzels Forlag.
Cohen, R. (2009). Resilience of the Internet to random breakdowns. Physical
review letters, 85(21): 4626-4628.
Csiernik, R,, Furze, P., Dromgole, L., & Rishchynski, G.M. (2008). Information Technology and Social Work—The Dark Side or Light Side. Journal of
Evidence-Based Social Work, 3(3-4): 9-25.
Cuadra, C.B. (2016). Socialtjänstens roll i samhällets krisberedskap – en studie ur personalens perspektiv [The role of the social services in societal crisis
preparedness – a study from the perspective of the staff]. Socialvetenskaplig
tidskrift, 23(2):129-150.
Cuadra, C.B. (2017b). Kunskapsunderlag för socialtjänstens delaktighet i kommunala risk- och sårbarhetsanalyser (Sorsa) [Knowledge base for the participation of the social services in municipal risk and vulnerability analyses (Sorsa)]. Research Report in Social Work 2017:2. Lund: Socialhögskolan, Lund University.
Cumming, S. Fitzpatrick, E., McAuliffe, D., McKain, S.. Martin, C., & Tonge, A. (2007). Raising the Titanic: Rescuing Social Work Documentation from the Sea of Ethical Risk. Australian Social Work, 60(2): 239-257.
27
De Rosa, E. (2017). Social innovation and ICT in social services: European experiences compared. The European Journal of Social Science Research, 30(4): 421–432.
Elliot, D., & Swartz, E. (2001). Business Continuity Management. A Crisis
Management Approach. New York and London: Routledge.
Fahlberg, G., & Larsson, M. (2016). Socialtjänstlagarna [The Social Services Acts]. 14th ed. Stockholm: Liber.
Fitch, D. (2015). A Conceptual Framework for Information Technology in Social Work Practice. Advances in Social Work, 16(1):15-30.
Galluch, P.S., Grover, V., & Thatcher, J. (2015). Interrupting the Workplace: Examining Stressors in an Information Technology Context. Journal of the
Association for Information Systems, 16(1):1-47.
Gillingham, P. (2016). Technology Configuring the User: Implications for the Redesign of Electronic Information Systems in Social Work. British Journal of
Social Work, 46(2): 323-338.
Hansen, A. (2015). Social Work Assessment Notes: A Comprehensive Documentation System. Health & Social Work, 40(3):191-200. Hedström, K (2004). Spår av datoriseringens värden – effekter av IT i
äldreomsorg [Traces of the values of computerisation – effects of IT in elderly
care]. Linköping Studies in Information Science, Dissertation No. 9.
Johansson, J., & Hassel, H. (2016) Beroendens betydelse i den sammankopplade samhället [The importance of dependencies in the connected society]. In S. Baez Ullberg & P. Becker (Eds.). Katastrofriskreducering. Perspektiv, praktik och
potential [Reduction of Disaster Risks. Perspectives, Practice and Potential].
Lund: Studentlitteratur.
Jönhill, J-I. (1997). Samhället som system och dess ekologiska omvärld. En studie
i Niklas Luhmanns sociologiska systemteori [Societies as systems and their
ecological environment. A study in Niklas Luhmann’s sociological systems theory]. Lund. Dissertation in Sociology 17.
28
Jönhill, J-I. (2012). Inklusion och exklusion. En distinktion som gör skillnad i det
mångkulturella samhället [Inclusion and exclusion. A distinction that makes a
difference in the multicultural society]. Malmö: Liber.
Luhmann, N. ([1984]1995). Social Systems. Stanford, CA: Stanford University Press.
Magrabi, F., Liaw, S.T., Arachi, D., et al. (2016). Identifying patient safety problems associated with information technology in general practice: an analysis of incident reports. BMJ Qual Saf 2016; 25:870–880.
Maturana, H.R., & Varela, F.J. (1980). Autopoiesis and Cognition. The
Realization of the Living. Dordrecht: D. Reidel Publishing Company.
Maturana, H.R., & Varela, F.J. (1987). The Tree of Knowledge. The Biological
Roots of Human Understanding. Massachusetts, MA: Shambhala Publications.
Michailakis, D., & Schirmer, W. (2017). Systemiska perspektiv på socialt arbete [Sustemic perspectives on social work]. Lund: Studentlitteratur.
Myndigheten för samhällsskydd och beredskap [Swedish Civil Contingencies Agency] (2011). Vägledning för Risk- och sårbarhetsanalyser [Guidelines for Risk and Vulnerability Analyses]. MSB245.
Myndigheten för samhällsskydd och beredskap [Swedish Civil Contingencies Agency] (2012). Kommunens informationssäkerhet – en vägledning [The Information Security of the Municipality – guidelines]. MSB508.
Myndigheten för samhällsskydd och beredskap [Swedish Civil Contingencies Agency] (2015b). Informationssäkerhet – trender 2015 [Information Security – Trends in 2015]. MSB779.
Myndigheten för samhällsskydd och beredskap [Swedish Civil Contingencies Agency] (2017). Nationell risk och förmågebedömning 2017 [National Risk and Ability Assessment 2017]. MSB1102.
National Board of Health and Welfare [Socialstyrelsen] (2014). Handbok för tillämpningen av bestämmelserna om lex Sarah [Handbook for the application of the regulations about Lex Sarah]. 2nd ed. Socialstyrelsen.
29
Perron, B.E. Taylor, H.O., Glass, J.E. Brown, G.W., & Margerum-Leys, J. (2010). Information and Communication Technologies in Social Work. Advances in
Social Work, 10(2): 67-81.
Reamer, F. G. (2005). Documentation in social work: evolving ethical and risk-management standards. Social Work, 50(4).
Reamer, F.G. (2013). Social Work in a Digital Age: Ethical and Risk Management Challenges. Social Work, 58(2): 163-172.
Roth, S., & Schütz, A. (2015). Ten Systems: Toward a Canon of Function Systems. Cybernetics and Human Knowing, 22(4): 11-31.
Schirmer, W., & Michailakis, D. (2015). The Luhmannian approach to exclusion/inclusion and its relevans to Social Work. Journal of Social Work, 15(1): 45-64.
SOSFS 2014:5 Dokumentation i verksamhet som bedrivs med stöd av SoL, LVU, LVM och LSS [Documentation in activities conducted with the support of SoL, LVU, LVM and LSS]. Socialstyrelsens författningssamling [The Code of Statutes of the National Board of Health and Welfare].
SOSFS 2011:9 Ledningssystem för systematiskt kvalitetsarbete [Management system for systematic quality work]. Socialstyrelsens författningssamling [The Code of Statutes of the National Board of Health and Welfare].
Spencer, L., Ritchie, J., Ormston, R., & O’Connor, W. (2014). Analysis: Principles and Practice. In J. Ritchie, J. Lewis, C. McNaughton Nicholls, & R. Ormston (Eds.). Qualitative Research Practice: A Guide for Social Science
