The Role of Social Network Technologies in Online Health Promotion: A Narrative Review of Theoretical and Empirical Factors Influencing Intervention Effectiveness

Background

Social network technologies have become part of health education and wider health promotion—either by design or happenstance. Social support, peer pressure, and information sharing in online communities may affect health behaviors. If there are positive and sustained effects, then social network technologies could increase the effectiveness and efficiency of many public health campaigns. Social media alone, however, may be insufficient to promote health. Furthermore, there may be unintended and potentially harmful consequences of inaccurate or misleading health information. Given these uncertainties, there is a need to understand and synthesize the evidence base for the use of online social networking as part of health promoting interventions to inform future research and practice.

Objective

Our aim was to review the research on the integration of expert-led health promotion interventions with online social networking in order to determine the extent to which the complementary benefits of each are understood and used. We asked, in particular, (1) How is effectiveness being measured and what are the specific problems in effecting health behavior change?, and (2) To what extent is the designated role of social networking grounded in theory?

Methods

The narrative synthesis approach to literature review was used to analyze the existing evidence. We searched the indexed scientific literature using keywords associated with health promotion and social networking. The papers included were only those making substantial study of both social networking and health promotion—either reporting the results of the intervention or detailing evidence-based plans. General papers about social networking and health were not included.

Results

The search identified 162 potentially relevant documents after review of titles and abstracts. Of these, 42 satisfied the inclusion criteria after full-text review. Six studies described randomized controlled trials (RCTs) evaluating the effectiveness of online social networking within health promotion interventions. Most of the trials investigated the value of a “social networking condition” in general and did not identify specific features that might play a role in effectiveness. Issues about the usability and level of uptake of interventions were more common among pilot studies, while observational studies showed positive evidence about the role of social support. A total of 20 papers showed the use of theory in the design of interventions, but authors evaluated effectiveness in only 10 papers.

Conclusions

More research is needed in this area to understand the actual effect of social network technologies on health promotion. More RCTs of greater length need to be conducted taking into account contextual factors such as patient characteristics and types of a social network technology. Also, more evidence is needed regarding the actual usability of online social networking and how different interface design elements may help or hinder behavior change and engagement. Moreover, it is crucial to investigate further the effect of theory on the effectiveness of this type of technology for health promotion. Research is needed linking theoretical grounding with observation and analysis of health promotion in online networks.

The Study Team for Early Life Asthma Research (STELAR) consortium 'Asthma e-lab': team science bringing data, methods and investigators together

We created Asthma e-Lab, a secure web-based research environment to support consistent recording, description and sharing of data, computational/statistical methods and emerging findings across the five UK birth cohorts. The e-Lab serves as a data repository for our unified dataset and provides the computational resources and a scientific social network to support collaborative research. All activities are transparent, and emerging findings are shared via the e-Lab, linked to explanations of analytical methods, thus enabling knowledge transfer. eLab facilitates the iterative interdisciplinary dialogue between clinicians, statisticians, computer scientists, mathematicians, geneticists and basic scientists, capturing collective thought behind the interpretations of findings.

Adoption of clinical decision support in multimorbidity: a systematic review

Background

Patients with multiple conditions have complex needs and are increasing in number as populations age. This multimorbidity is one of the greatest challenges facing health care. Having more than 1 condition generates (1) interactions between pathologies, (2) duplication of tests, (3) difficulties in adhering to often conflicting clinical practice guidelines, (4) obstacles in the continuity of care, (5) confusing self-management information, and (6) medication errors. In this context, clinical decision support (CDS) systems need to be able to handle realistic complexity and minimize iatrogenic risks.

Objective

The aim of this review was to identify to what extent CDS is adopted in multimorbidity.

Methods

This review followed PRISMA guidance and adopted a multidisciplinary approach. Scopus and PubMed searches were performed by combining terms from 3 different thesauri containing synonyms for (1) multimorbidity and comorbidity, (2) polypharmacy, and (3) CDS. The relevant articles were identified by examining the titles and abstracts. The full text of selected/relevant articles was analyzed in-depth. For articles appropriate for this review, data were collected on clinical tasks, diseases, decision maker, methods, data input context, user interface considerations, and evaluation of effectiveness.

Results

A total of 50 articles were selected for the full in-depth analysis and 20 studies were included in the final review. Medication (n=10) and clinical guidance (n=8) were the predominant clinical tasks. Four studies focused on merging concurrent clinical practice guidelines. A total of 17 articles reported their CDS systems were knowledge-based. Most articles reviewed considered patients’ clinical records (n=19), clinical practice guidelines (n=12), and clinicians’ knowledge (n=10) as contextual input data. The most frequent diseases mentioned were cardiovascular (n=9) and diabetes mellitus (n=5). In all, 12 articles mentioned generalist doctor(s) as the decision maker(s). For articles reviewed, there were no studies referring to the active involvement of the patient in the decision-making process or to patient self-management. None of the articles reviewed adopted mobile technologies. There were no rigorous evaluations of usability or effectiveness of the CDS systems reported.

Conclusions

This review shows that multimorbidity is underinvestigated in the informatics of supporting clinical decisions. CDS interventions that systematize clinical practice guidelines without considering the interactions of different conditions and care processes may lead to unhelpful or harmful clinical actions. To improve patient safety in multimorbidity, there is a need for more evidence about how both conditions and care processes interact. The data needed to build this evidence base exist in many electronic health record systems and are underused.

Simulating realistic enough patient records

Information systems for storing, managing and manipulating electronic medical records must place an emphasis on maintaining the privacy and security of those records. Though the design, development and testing of such systems also requires the use of data, the developers of these systems, rarely also their final end users, are unlikely to have ethical or governance approval to use real data. Alternative test data is commonly either randomly produced or taken from carefully anonymised subsets of records. In both cases there are potential shortcomings that can impact on the quality of the product being developed. We have addressed these shortcomings with a tool and methodology for efficiently simulating large amounts of realistic enough electronic patient records which can underpin the development of data-centric electronic healthcare systems.

Mobile early detection and connected intervention to coproduce better care in severe mental illness

Current approaches to the management of severe mental illness have four major limitations: 1) symptom reporting is intermittent and subject to problems with reliability; 2) service users report feelings of disengagement from their care planning; 3) late detection of symptoms delay interventions and increase the risk of relapse; and 4) care systems are held back by the costs of unscheduled hospital admissions that could have been avoided with earlier detection and intervention. The ClinTouch system was developed to close the loop between service users and health professionals. ClinTouch is an end-to-end secure platform, providing a validated mobile assessment technology, a web interface to view symptom data and a clinical algorithm to detect risk of relapse. ClinTouch integrates high-resolution, continuous longitudinal symptom data into mental health care services and presents it in a form that is easy to use for targeting care where it is needed. The architecture and methodology can be easily extended to other clinical domains, where the paradigm of targeted clinical interventions, triggered by the early detection of decline, can improve health outcomes.

Using String Metrics to Identify Patient Journeys through Care Pathways

Given a computerized representation of a care pathway and an electronic record of a patient's clinical journey, with potential omissions, insertions, discontinuities and reordering, we show that we can accurately match the journey to a particular route through the pathway by converting the problem into a string matching one. We discover that normalized string metrics lead to more unique pathway matches than non-normalized string metrics and should therefore be given preference when using these techniques.

Making audit actionable: an example algorithm for blood pressure management in chronic kidney disease

Despite widespread use of clinical guidelines, actual care often falls short of ideal standards. Electronic health records (EHR) can be analyzed to provide information on how to improve care, but this is seldom done in sufficient detail to guide specific action. We developed an algorithm to provide practical, actionable information for care quality improvement using blood pressure (BP) management in chronic kidney disease (CKD) as an exemplar. We used UK clinical guidelines and EHR data from 440 patients in Salford (UK) to develop the algorithm. We then applied it to 532,409 individual patient records, identifying 11,097 CKD patients, 3,766 (34%) of which showed room for improvement in their care: either through medication optimization or better BP monitoring. Manual record reviews to evaluate accuracy indicated a positive-predictive value of 90%. Such algorithms could help improve the management of chronic conditions by providing the missing link between clinical audit and decision support.

Breast cancer risk in young women in the national breast screening programme: implications for applying NICE guidelines for additional screening and chemoprevention

In the United Kingdom, women at moderate and high risk of breast cancer between the ages of 40 and 49 years are eligible for annual mammographic screening and preventive therapy with tamoxifen. Here, we estimate the numbers of women in a population eligible for this service and the proportion of breast cancers detected in this group compared with the whole population. Women <50 attending for mammographic screening in the National Health Service Breast Screening Programme (NHSBSP) completed a risk questionnaire. The proportion at moderate and high risk according to National Institute of Health Care Excellence (NICE) guidelines was estimated. An estimate was also made using a different model of risk estimation (Tyrer-Cuzick). The numbers of cancers detected in the moderate/high risk groups were compared with numbers detected in the whole population. Completed questionnaires were available for 4,360 women between ages 46 and 49 years. Thirty women [0.7%; 95% confidence interval (CI), 0.5-1.0%] were at high risk and 130 (3.0%, 2.5-3.5%) were at moderate risk according to NICE guidelines. Thirty-seven cancers were detected by mammography in the whole group. Five of these were found in the moderate-/high-risk group giving a 3.2-fold increase in detection compared with the standard risk group. More women were assigned to the moderate- or high-risk group using the Tyrer-Cuzick model (N = 384), but the numbers of cancers in this group were not appreciably increased (N = 8). Systematic assessment of family history in primary care or through population-based screening will identify appreciable numbers of women in their forties, eligible for additional surveillance and chemoprevention.

The temporal association between self-injurious thoughts and psychotic symptoms: a mobile phone assessment study

The relationship between psychotic symptoms and self-injurious thoughts (SITs) remains unclear. The short-term temporal associations between psychotic symptoms and SITs were explored. A sample of 36 people with a diagnosis of a psychotic disorder or at-risk mental state completed mobile phone-based measures at multiple times each day for 1 week. Clustered regression with time-lagged variables supported a relationship between paranoia and subsequent SITs. Hallucinations did not predict these thoughts when controlling for paranoia. The role of specific psychotic symptoms in triggering SITs is highlighted and the importance of considering these factors in risk management is discussed.

A comparison of two delivery modalities of a mobile phone-based assessment for serious mental illness: native smartphone application vs text-messaging only implementations

BACKGROUND

Mobile phone-based assessment may represent a cost-effective and clinically effective method of monitoring psychotic symptoms in real-time. There are several software options, including the use of native smartphone applications and text messages (short message service, SMS). Little is known about the strengths and limitations of these two approaches in monitoring symptoms in individuals with serious mental illness.

OBJECTIVE

The objective of this study was to compare two different delivery modalities of the same diagnostic assessment for individuals with non-affective psychosis-a native smartphone application employing a graphical, touch user interface against an SMS text-only implementation. The overall hypothesis of the study was that patient participants with sewrious mental illness would find both delivery modalities feasible and acceptable to use, measured by the quantitative post-assessment feedback questionnaire scores, the number of data points completed, and the time taken to complete the assessment. It was also predicted that a native smartphone application would (1) yield a greater number of data points, (2) take less time, and (3) be more positively appraised by patient participant users than the text-based system.

METHODS

A randomized repeated measures crossover design was employed. Participants with currently treated Diagnostic and Statistical Manual (Fourth Edition) schizophrenia or related disorders (n=24) were randomly allocated to completing 6 days of assessment (four sets of questions per day) with a native smartphone application or the SMS text-only implementation. There was then a 1-week break before completing a further 6 days with the alternative delivery modality. Quantitative feedback questionnaires were administered at the end of each period of sampling.

RESULTS

A greater proportion of data points were completed with the native smartphone application in comparison to the SMS text-only implementation (β = -.25, SE=.11, P=.02), which also took significantly less time to complete (β =.78, SE= .09, P<.001). Although there were no significant differences in participants' quantitative feedback for the two delivery modalities, most participants reported preferring the native smartphone application (67%; n=16) and found it easier to use (71%; n=16). 33% of participants reported that they would be willing to complete mobile phone assessment for 5 weeks or longer.

CONCLUSIONS

Native smartphone applications and SMS text are both valuable methods of delivering real-time assessment in individuals with schizophrenia. However, a more streamlined graphical user interface may lead to better compliance and shorter entry times. Further research is needed to test the efficacy of this technology within clinical services, to assess validity over longer periods of time and when delivered on patients' own phones.

Septo-optic dysplasia: antenatal risk factors and clinical features in a regional study

BACKGROUND

Septo-optic dysplasia (SOD) is a disorder with postulated environmental and genetic aetiology. This study delineates clinical features and potential perinatal environmental factors along with epidemiology in SOD children.

METHODS

Assessment of patients with SOD triad features in the UK West Midlands region.

RESULTS

Of 227 patients identified between 1998 and 2009 with 1 or more feature of the triad, 55 had midline defects, 149 had optic nerve hypoplasia and 132 had hypopituitarism. Eighty-eight children (52% males; incidence 8.3/100,000 live births) had SOD defined as 2 out of 3 features and 21 (24%) had all 3. Sixty-one percent had anterior pituitary deficiency and 21.5% had diabetes insipidus. Median maternal/paternal ages in SOD were 21 and 23.5 years, compared to UK means of 29.3 and 32.4 years (p < 0.001). First trimester bleeding was markedly increased at 12/48 (25%) compared to 0.07% in the UK (p < 0.001). Ethnicity showed a non-significant higher prevalence in Afro-Caribbean and mixed race groups, and significantly lower prevalence (p = 0.004) in South Asian groups compared to West Midland and Birmingham city data: 8% versus 2.5 and 6.7%, 9% versus 1.8 and 3.2% and 3% versus 8.4 and 21%, respectively.

CONCLUSIONS

SOD is associated with younger maternal and paternal age, primigravida births and ethnic differences. Increased first trimester bleeding may indicate that SOD is a vascular disruption sequence.

Integrating mobile-phone based assessment for psychosis into people's everyday lives and clinical care: a qualitative study

BACKGROUND

Over the past decade policy makers have emphasised the importance of healthcare technology in the management of long-term conditions. Mobile-phone based assessment may be one method of facilitating clinically- and cost-effective intervention, and increasing the autonomy and independence of service users. Recently, text-message and smartphone interfaces have been developed for the real-time assessment of symptoms in individuals with schizophrenia. Little is currently understood about patients' perceptions of these systems, and how they might be implemented into their everyday routine and clinical care.

METHOD

24 community based individuals with non-affective psychosis completed a randomised repeated-measure cross-over design study, where they filled in self-report questions about their symptoms via text-messages on their own phone, or via a purpose designed software application for Android smartphones, for six days. Qualitative interviews were conducted in order to explore participants' perceptions and experiences of the devices, and thematic analysis was used to analyse the data.

RESULTS

Three themes emerged from the data: i) the appeal of usability and familiarity, ii) acceptability, validity and integration into domestic routines, and iii) perceived impact on clinical care. Although participants generally found the technology non-stigmatising and well integrated into their everyday activities, the repetitiveness of the questions was identified as a likely barrier to long-term adoption. Potential benefits to the quality of care received were seen in terms of assisting clinicians, faster and more efficient data exchange, and aiding patient-clinician communication. However, patients often failed to see the relevance of the systems to their personal situations, and emphasised the threat to the person centred element of their care.

CONCLUSIONS

The feedback presented in this paper suggests that patients are conscious of the benefits that mobile-phone based assessment could bring to clinical care, and that the technology can be successfully integrated into everyday routine. However, it also suggests that it is important to demonstrate to patients the personal, as well as theoretical, benefits of the technology. In the future it will be important to establish whether clinical practitioners are able to use this technology as part of a personalised mental health regime.

e-Labs and the stock of health method for simulating health policies

Regional outcomes of national health policies are difficult to forecast. This is partly due to a lack of realistically complex models that can be used to appraise policy options and partly a lack of accessible and adaptable tools that can be used to simulate the consequences of policy decisions. These barriers might be overcome by exploiting the commoditization of massively parallel computing architectures, advances in machine learning, and the increased availability of large-scale linked healthcare data. This paper presents a novel modelling methodology, The Stock of Health, for harnessing emerging data and computational resources to simulate health policy, with application initially to coronary heart disease. We detail the use of multi-core graphical processing architectures to facilitate a micro-simulation approach. The simulation tools have been deployed through the IMPACT Framework. We explore how this framework can be extended to support the sharing and reuse of policy models and simulations based on the digital publishing concept of e-Lab.

Affective instability prior to and after thoughts about self-injury in individuals with and at-risk of psychosis: a mobile phone based study

It has been proposed that affective instability may be associated with thoughts about self-injury. The aim of this study was to test the hypotheses that instability in feelings of depression, but not anxiety, guilt, or hostility, would predict greater concurrent and subsequent thoughts about self-injury. Thirty-six individuals with psychosis completed questions on touch-screen mobile phones at semi-random times each day for one week. The instability of depression predicted greater concurrent and subsequent levels of thoughts about self-injury, even when controlling for depression level. Conversely, self-injurious thoughts predicted more stable depression. The instability of guilt, anxiety, and hostility did not significantly predict levels of thoughts about self-injury. Results indicate that a variable depressive state may trigger the onset of thoughts about self-injury, which increases the risk of its subsequent recurrence. The onset of self-injurious thoughts may, however, have a stabilizing effect on subsequent depression.

Missed opportunities mapping: computable healthcare quality improvement

INTRODUCTION

Analysing variance from care pathways in situations when adverse health outcomes have occurred may identify missed opportunities for healthcare improvement.

METHODS

We developed a computational model for contrasting observed with expected care in pathway searches of coded electronic health records (EHRs). The model was applied in Salford, UK, looking at blood pressure (BP) control and cardiovascular disease (CVD) events. BP was summarised as the integral of serial measurements.

RESULTS

A missed opportunities mapping (MOM) model consisting of a collection of disease Events and pathophysiologic States was used to articulate all CVD scenarios conceived. In 3718 patients suffering CVD events in Salford (2007-2012), 1186 (32%) had suboptimal BP control. This missed opportunity detection rose to 36% using the integral instead of the most recent BP record.

CONCLUSIONS

MOM provides a useful, computable model for encoding care pathways and searching EHRs to detect variations from expected care. Further research is needed in other disease areas. The indications however, are that this model could be used to embed healthcare quality improvement at both patient and population levels.

Verbal protocols for assessing the usability of clinical decision support: the retrospective sense making protocol

We compare the effectiveness of two types of verbal protocol, concurrent think aloud vs. retrospective sense making, for evaluating the usability of a clinical decision support tool. Thirty-five medical and nursing students participated in a usability experiment. Participants were asked to complete seven tasks using the system under evaluation. Eighteen students were allocated to the concurrent think aloud group and the remainder followed the retrospective protocol. The retrospective protocol was significantly more sensitive than the concurrent protocol in recording unique usability problems related to users' cognitive behaviour. These problems concerned the interpretation and comprehension of statistical output, search results and system messages. These findings can be explained by the retrospective protocol's greater ability to detect compound usability problems, capturing the cognitive dimensions of users' interactions with the interface in greater depth. Evaluations of clinical decision support systems should take an evidence-based approach to selecting protocols.

The feasibility and validity of ambulatory self-report of psychotic symptoms using a smartphone software application

BACKGROUND

Semi-structured interview scales for psychosis are the gold standard approach to assessing psychotic and other symptoms. However, such assessments have limitations such as recall bias, averaging, insensitivity to change and variable interrater reliability. Ambulant, real-time self-report assessment devices may hold advantages over interview measures, but it needs to be shown that the data thus collected are valid, and the collection method is acceptable, feasible and safe. We report on a monitoring system for the assessment of psychosis using smartphone technology. The primary aims were to: i) assess validity through correlations of item responses with those on widely accepted interview assessments of psychosis, and ii) examine compliance to the procedure in individuals with psychosis of varying severity.

METHODS

A total of 44 participants (acute or remitted DSM-4 schizophrenia and related disorders, and prodromal) completed 14 branching self-report items concerning key psychotic symptoms on a touch-screen mobile phone when prompted by an alarm at six pseudo-random times, each day, for one week. Face to face PANSS and CDS interviews were conducted before and after the assessment period blind to the ambulant data.

RESULTS

Compliance as defined by completion of at least 33% of all possible data-points over seven days was 82%. In the 36 compliant participants, 5 items (delusions, hallucinations, suspiciousness, anxiety, hopelessness) showed moderate to strong (rho 0.6-0.8) associations with corresponding items from interview rating scales. Four items showed no significant correlation with rating scales: each was an item based on observable behaviour. Ambulant ratings showed excellent test-retest reliability and sensitivity to change.

CONCLUSIONS

Ambulatory monitoring of symptoms several times daily using smartphone software applications represents a feasible and valid way of assessing psychotic phenomena for research and clinical management purposes. Further evaluation required over longer assessment periods, in clinical trials and service settings.

Active assistance technology for health-related behavior change: an interdisciplinary review

Background

Information technology can help individuals to change their health behaviors. This is due to its potential for dynamic and unbiased information processing enabling users to monitor their own progress and be informed about risks and opportunities specific to evolving contexts and motivations. However, in many behavior change interventions, information technology is underused by treating it as a passive medium focused on efficient transmission of information and a positive user experience.

Objective

To conduct an interdisciplinary literature review to determine the extent to which the active technological capabilities of dynamic and adaptive information processing are being applied in behavior change interventions and to identify their role in these interventions.

Methods

We defined key categories of active technology such as semantic information processing, pattern recognition, and adaptation. We conducted the literature search using keywords derived from the categories and included studies that indicated a significant role for an active technology in health-related behavior change. In the data extraction, we looked specifically for the following technology roles: (1) dynamic adaptive tailoring of messages depending on context, (2) interactive education, (3) support for client self-monitoring of behavior change progress, and (4) novel ways in which interventions are grounded in behavior change theories using active technology.

Results

The search returned 228 potentially relevant articles, of which 41 satisfied the inclusion criteria. We found that significant research was focused on dialog systems, embodied conversational agents, and activity recognition. The most covered health topic was physical activity. The majority of the studies were early-stage research. Only 6 were randomized controlled trials, of which 4 were positive for behavior change and 5 were positive for acceptability. Empathy and relational behavior were significant research themes in dialog systems for behavior change, with many pilot studies showing a preference for those features. We found few studies that focused on interactive education (3 studies) and self-monitoring (2 studies). Some recent research is emerging in dynamic tailoring (15 studies) and theoretically grounded ontologies for automated semantic processing (4 studies).

Conclusions

The potential capabilities and risks of active assistance technologies are not being fully explored in most current behavior change research. Designers of health behavior interventions need to consider the relevant informatics methods and algorithms more fully. There is also a need to analyze the possibilities that can result from interaction between different technology components. This requires deep interdisciplinary collaboration, for example, between health psychology, computer science, health informatics, cognitive science, and educational methodology.

eLab: bringing together people, data and methods to enhance knowledge discovery in healthcare settings

The discovery of knowledge from raw data is a multistage process, that typical requires collaboration between experts from disparate disciplines, and the application of a range of methods tailored to the research question. The aim of the eLab is to provide a web-based environment for health professionals and researchers to access health datasets, share knowledge and expertise and to apply methods for analysis and visualization of the results. The eLab is built around the core concept of the Research Object as the mechanism for preserving, reusing and disseminating the knowledge discovery process. The possible range of applications of the eLab is vast, and so the consideration of the trade off between specificity and generality is an important one, that is reflected in the requirements. The architecture and implementation of the eLab is described, and we report on the deployment of eLabs for applications in primary care, long-term conditions management, bariatric surgery and public health.

COCPIT: a tool for integrated care pathway variance analysis

Electronic Health Record (EHR) data has the potential to track patients' journeys through healthcare systems. Many of those journeys are supposed to follow Integrated Care Pathways (ICPs) built on evidence based guidelines. An ICP for a particular condition sets out "what should happen", whereas the EHR records "what did happen". Variance analysis is the process by which the difference between expected and actual care is identified. By performing variance analysis over multiple patients, patterns of deviation from idealised care are revealed. The use of ICP variance analysis, however, is not as widespread as it could be in healthcare quality improvement processes - we argue that this is due to the difficulty of combining the required specialist knowledge and skills from different disciplines. COCPIT (Collaborative Online Care Pathway Investigation Tool) was developed to overcome this difficulty and provides clinicians and health service managers with a web-based tool for Care Pathway Variance Analysis.