PERSONAE: Developing Anticoagulation Self-Report Measure in Adults
Study Details
Study Description
Brief Summary
The use of oral anticoagulants has been increasing globally to prevent strokes associated with atrial fibrillation and heart valve replacements. Patients on long-term anticoagulation have the option to self-test their blood clotting time using point-of-care testing equipment (POCT). They can choose to self-manage their medication dosage or self-monitor and seek dose adjustments from a clinic. Studies have shown that self-testing approaches are effective and cost-effective. However, future research should focus on understanding the factors that influence people's preparedness for self-testing, such as health literacy, self-efficacy, and perceived support. Assessing preparedness is crucial for tailoring healthcare delivery and supporting patient education. Highlighting preparedness can also encourage the adoption of self-testing approaches in countries facing barriers to implementation.
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
Detailed Description
Scientific Rationale / Background
In every situation where the direct oral anticoagulants are not safe (e.g., adults with multiple morbidities), optimal anticoagulation with warfarin or other vitamin K antagonists (VKAs) like acenocoumarol or phenprocoumon could potentially prevent more than half of the strokes associated with atrial fibrillation and heart valve replacements with a relatively low risk of major bleeding complications (Garcia et al., 2022). Globally, the proportion of patients taking oral anticoagulants has been rising (Hsu et al., 2022). For instance, roughly 2 out of 100 Italians are taking oral anticoagulants (both direct and indirect drugs) (Marchetti et al., 2020). As a result, regular monitoring of prothrombin time and dose adjustment by a specialized hospital service, primary care physician, registered nurse, nurse practitioner, or pharmacist is necessary for people in treatment with VKAs.
People using long-term oral anticoagulation could currently use point-of-care testing equipment (POCT) to keep track of their blood clotting time, which is represented by the international normalized ratio (INR) (Kantito et al., 2018). Patients who self-test have two options: they can self-manage (change their medication dosage following a predetermined dose-INR schedule) or self-monitor (contact a clinic and ask for the proper dose adjustment) (Kuljis et al., 2017). According to several published studies and systematic reviews, these approaches to monitoring anticoagulant medication were found to be as good as or better than routine monitoring by the traditional physician-managed approach (Heneghan et al., 2016; Magon et al., 2020).
The quality of anticoagulation control is traditionally expressed by the time in the therapeutic range (%TTR). More time spent by patients within the therapeutic range (higher %TTR) leads to a decline in thromboembolic events with no corresponding rise in risks. In this context, CoaguChek XS and other coagulometers are presently recommended for patients with atrial fibrillation and heart valve disease by the National Institute for Health and Care Excellence (NICE) advice in the UK (NICE diagnostics guidance [DG14]) (NICE, 2016). Similar recommendations are available worldwide, which were triggered by the COVID-19 pandemic (Kow et al., 2020).
Although self-monitoring and self-management (self-testing approaches) are supported by a good level of evidence showing their effectiveness and cost-effectiveness, the last Cochrane review on this topic concluded that future research should examine the factors defining the preparedness of people's decisions for self-testing (or not) and should take into account consumer knowledge of self-management, factors that lead people to seek medical attention and detailed medical information (health literacy), self-efficacy or self-confidence in one's ability to self-manage, and perceived or actual support. Assessing the preparedness for self-management and self-monitoring is highly strategic to tailoring the healthcare delivery, boosting its efficiency, and supporting an adequate educational preparation of suitable patients for self-testing (Magon et al., 2016). Highlighting the preparedness for self-management and self-monitoring could also be a strategy to boost the need for adopting self-testing approaches in those countries, such as Italy, where the rate of adoption of this approach is undermined by several organizational and professional barriers, such as the costs of POCT and the unjustified fears of losing a direct control on how patients manage their treatment (Magon et al., 2022).
Study design
Following the recommendations to develop self-reported scales (Rattray & Jones, 2007), a methodological, multiphase study comprises two main phases.
Phase one will focus on conceptualizing the PERSONAE scale and will be based on a developmental methodological approach, including three primary steps. The primary step will be a literature review to include all the elements linked to the preparedness for self-monitoring and self-management of oral anticoagulation. The second step will be a consensus meeting among developers to examine the literature review findings and endorse its synthesis. Then, the third step will aim to operationalize the situations, and an external panel of experts will be involved to determine the face and content validity of the newly developed scale.
Phase two will consist of the validation procedures. Initially, the PERSONAE scale showing items with acceptable content validity will be used for cross-sectional data collection. This specific step will determine, using the Mokken scaling analysis (MSA) procedure, how the items in the first version of the PERSONAE scale will behave in response to varying levels of the theoretical latent trait (preparedness) (Watson et al., 2012). The MSA is chosen by design in this study over the exploratory fact analysis because it investigates the behaviors of individual items about a single latent characteristic rather than looking at the link between items, allowing researchers to select the items with the highest scalability toward a brief, unidimensional measure that can be computed in a single score. In other words, by adding just the most pertinent questions established using a hierarchical method, MSA satisfied the aim of establishing a quick assessment of the preparedness for self-monitoring and self-management of oral anticoagulation. Once a stable unidimensional PERSONAE scale is developed, a second cross-sectional data collection will be performed to cross-validate the plausible unidimensional structure of the scale.
Participants and procedures
In phase one, a panel of 8-12 experts (i.e., nurses, physicians, and pharmacists) will assess the relevance and clarity of the proposed items derived from a literature review performed by the investigators (content validity). The items will represent the situations reflecting how much patients feel to be prepared to safely and effectively start self-monitoring and self-management. The involved experts will sign a written consent form to allow investigators to use their socio-demographic and professional characteristics in the scientific reporting because content validity is closely linked to the characteristics of the involved experts. In other words, this phase cannot be anonymous to allow investigators to increase the transparency regarding the characteristics of the panel by collecting the following variables: sex, age, academic title, previous experience in studies aimed to test content validity, and experience in oral anticoagulation.
Once adequate content validity is obtained in phase two, the study's coordinating center, IRCCS Policlinico San Donato, will start the data collection. This protocol encompasses two cross-sectional data collection rounds by employing web surveys to collect data by ensuring anonymity. A convenience sampling procedure will be employed by using the webpages and social network media pages of associations involving anticoagulated patients, such as the "Federazione delle Associazioni Italiane dei Pazienti Anticoagulati" (Feder-AIPA) and "Associazione Italiana Cardiopatici Congeniti" (AICCA). Once the first data collection round is completed (i.e., when sample size will be achieved), the selection of the participants for the second round of data collection will be based on the same approach as the first by adding a communication aimed to explain that if a patient has already answered to the survey, it is not possible to re-fill the survey for the second data collection round for avoiding overlapping responses.
Sampling procedure
Data collection will be performed using a cloud-data recording system and involving a web survey disseminated by the promoter using social media campaigns, newsletters, and flyers. The sampling procedure will be managed by the local study coordinator and the researcher from the team accountable for managing the procedure following good clinical practice principles. As per previous studies, a response rate of approximately 60% of the invited patients is expected (Magon et al., 2020).
Sample size
Considering that the number of items in the PERSONAE scale will depend on the literature review and the content validity phase, few elements in this pre-development stage are available to run precise Monte Carlo simulations for determining the required sample size to achieve at least a power of 0.80 in rejecting the null hypothesis that total-scale H coefficient = 0. To determine the sample size needed to perform the MSA, the study by Straat et al. (2014) showed that a sample of 500 participants might be adequate to perform MSA. Similar results have been shown in the Monte Carlo simulation performed by Ligtvoet et al. (2010). For this reason, a sample size of 500 participants is required for the first round of data collection. The MSA procedure will allow reducing, based on a specific hierarchy, the number of items, including only the ones that will fulfill the monotonicity and the invariant item ordering assumptions; the first 10-15 items resulting from the selection will define the preparedness for self-monitoring and self-management of oral anticoagulation in adult patients.
For this reason, the required sample for the second round of data collection was based on Monte Carlo simulations performed in Mplus 8.1 (Muthén & Muthén, 2017), considering several possibilities (normality vs. non-normal distribution; missing values lower than 5% vs. missing values higher than 5%). With no missing data and normally distributed variables, the smallest sample size for performing CFA requires 150 patients to have a power of 0.80 in rejecting the null hypothesis that the factor correlation is zero. When the likelihood of missing data is added to the simulation model, a sample size of 175 is needed to achieve a power of 0.80. For a power of 0.80 when using the CFA model with non-normal items and no missing data, 245 patients are required. While adding the likelihood of missing data, a sample size of 305 patients is required for a power of 0.80 in rejecting the null hypothesis that the factor correlation is zero. The last step in the simulation provides the most conservative approach and, for this reason, represents the sample size for the second round of data collection. For each simulation, data were generated from a population with hypothesized parameter values derived from previous studies with items exploring potential proxies of preparedness, such as knowledge regarding AVKs, health literacy, and health engagement (Magon et al., 2018, 2021).
Eligibility criteria
In phase one, inclusion criteria for experts will be related to a proven background in managing patients with VKAs (at least one year), at least a master of science degree in nursing, pharmacy, or an MD degree, and Italian language proficiency. The exclusion criterium will be related to the presence of conflicts of interest regarding the topic.
For phase two, inclusion criteria are detailed in a subsequest table ("Groups and Interventions" section).
Data collection tools
Data collection will be performed through the version of the PERSONAE scale derived from the content validity phase, a socio-demographic and clinical data form, and a graphic rating scale based on 5-point assessments (from 0 = strongly disagree to 5 = completely agree) to assess how patients agree in stating that they would like to receive an education using a list of technology-assisted learning options (each option will be rated).
The socio-demographic data will be age (years), sex (male, female, other), marital status (married, unmarried), educational level (lower than secondary school diploma, secondary school diploma, university), and occupation (active worker, non-active worker). Clinical variables will be the time in oral anticoagulation, clinical indication for the anticoagulation, time in therapeutic range (TTR), and anamnesis of thromboembolic or bleeding complications in the last three months.
Data analysis
The package "mokken" from the statistical program R (R Foundation for Statistical Computing, Vienna, Austria) will be employed for performing MSA in the sample derived from the first data collection round (Ark et al., 2021). The MSA allows developers to evaluate each item hierarchically and select the most relevant ones to evaluate the preparedness (the first 10-15 items), which is the focus of the PERSONAE scale. This approach assumes that a score on one item is solely a function of the latent trait (preparedness) level and not a score on any other items. The automated item selection procedure (AISP) will be used in the first stage of the MSA to determine how many possible scales will emerge from the provided data. The scalability of the items will be assessed by evaluating the Loevinger's coefficient of homogeneity (coefH) function and then checking for monotonicity violations after the items for the most likely scale measuring the preparedness for self-monitoring and self-management of oral anticoagulation were identified (only violations below 80 are consistent with the requirements of a Mokken scale). Loevinger's coefficients of homogeneity and their confidence intervals (CI) at the scale level (Hs), item level (scalability of an individual item, Hi), and pairs of items (HiJ) will be computed. H values equal to or greater than 0.30, 0.40, or 0.50, respectively, show weak, moderate, or strong scales. The inspection of item pairs will sustain the exclusion of those violating monotonicity and also provide an assessment of floor and ceiling effects in data distribution. The appropriate items will then be evaluated using invariant item ordering (IIO), which includes a backward item selection method algorithm to narrow the pool of questions to those that best fit the requirements of the used hierarchical ordering of respondents based on monotone homogeneity. The IIO procedure will remove items one at a time until no significant violations are found. For each succeeding stage in which the algorithm will be used, the Htrans (HT) coefficient will express the accuracy of the IIO selection. HT greater than 0.30 is deemed sufficient to denote proper IIO process and consistent Mokken scales (Watson et al., 2012). The final set of questions will be used to calculate the rho coefficients computed using the Molenaar-Sijtsma technique for reliability.
Confirmatory factor analysis will be employed to cross-validate the plausible unidimensional structure of the PERSONAE scale derived from the MSA (theoretically including 10-15 items) using Mplus 8.1 (Muthén & Muthén, 2017). The first model will be unrestricted and unspecified based on a maximum likelihood robust estimation of the parameters. A confirmatory factor analysis with covariates (MIMIC) to estimate the measurement invariance between subgroups if linear relationships between the characteristics of participants and the scale's total score will be detected. In the case of employing MIMIC models, the amount of reduced χ2 for each parameter will be explored using the "modification indices" command that will allow authors to detect which items might cause such violation of invariance (because showing different behaviors between groups). Items causing invariance will be evaluated for possible deletion from the final scale. The suitability of confirmatory factor analyses in explaining sample statistics will be evaluated using the following indices: χ2, χ2/degree of freedom (DF, acceptable range: 1.5-5), the comparative fit index (CFI), and Tucker-Lewis index (TLI) (CFI and TLI > 0.90 indicate a good fit), and the root mean square error of approximation (RMSEA; values of 0.05 or less indicating a well-fitting model, from 0.05 to 0.08 indicating an adequate fit, and 0.10 or greater indicating under-optimal fit).
Missing data will be managed using an available-case approach for missingness lower than 5% or regression imputations for missingness equal to or higher than 5% if the missing at random assumptions will be empirically met. Analytics will be performed with type I error equal to 5% and using Mplus, version 8.1 (Los Angeles, CA, USA), IBM SPSS® Statistics for Microsoft Windows, version 27 (IBM Corp., Armonk, NY, USA), and R 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/).
Ethical considerations
The first stage of the study (content validity) will involve experts who will be asked to sign a written informed consent. The data collection via web surveys designed in the second part of the study can be defined as anonymous because it is impossible to associate any specific sample with a specific individual using the data collected. According to current law (European Regulation 2016/679, Legislative Decree 101/2018), using a type of written informed consent for this type of study is unnecessary because it would compromise the study's anonymity. On the other hand, it is recommended that patients be fully informed with a presentation page that is simple to read and describes the goal of the survey before they complete the online questionnaire.
Implications for clinical practice
This project will allow clinicians (nurses, hematologists, general practitioners, and cardiologists) to assess the preparedness for self-management and self-monitoring in patients treated using VKAs. Assessing their preparedness could have several implications for clinical practice.
First, in Italy and several countries with a similar universal healthcare system, the utilization of POCT is under-optimal (Magon et al., 2022), and highlighting a structured path to best prepare patients for self-testing approaches might optimize the practical adoption of these safe, effective, and cost-effective managing strategies (self-monitoring and self-management). This aspect is particularly meaningful as it was demonstrated that potentially eligible patients for self-testing managed with a traditional approach report lower levels of mental health because they feel unable to express their potential contribution to managing their condition (Magon et al., 2020).
Second, the preparedness for self-management and self-monitoring in patients treated using VKAs might guide the designing of tailored educational activities for patients. The need for tailored education in patients with VKAs has been extensively proposed in the literature, especially with the restrictions introduced by the COVID-19 pandemic (Price & Ansell, 2022). These educational approaches should also drive interprofessional collaboration and shared decision-making (Gee et al., 2022).
Third, the secondary aims of this project will provide an initial epidemiological description of the preparedness in the Italian setting and can be used to prioritize public health policies to close the possible emerging gaps. Furthermore, the description of the most appreciated technology-assisted learning options by patients could be informative for clinicians and public health officers.
Study Design
Arms and Interventions
| Arm | Intervention/Treatment |
|---|---|
| Sample derived from the first data collection round (MSA) Patients have to be aged 18 years and older, cross in an anticoagulation clinic for at least three months, have appropriate cognitive functioning as determined by the six-item screener, and have the ability to give informed consent will be the inclusion criteria. Patients will be excluded if their Charlson Comorbidity Index (CCI) is high (CCI > 4) to account for any comorbidity-related variability and assure the homogeneity of the study sample. |
Other: Participation in an web-based survey
Patients who agree to participate will access a web-based survey (SurveyMonkey) via social media, mobile messages, or emails. The invitation will provide study details, eligibility criteria, and tasks. Patients will confirm inclusion/exclusion criteria upon access, and those ineligible will opt out. Data collection will take approximately 10 minutes per patient. Collected data will be sent to an electronic data collector and saved as an eCRF file in REDCap. The study will conduct two cross-sectional data collection rounds using web surveys to ensure anonymity. Convenience sampling will target anticoagulated patient associations' webpages and social media (e.g., Feder-AIPA, AICCA). In the second round, participants will be informed not to resubmit if they have already completed the survey to avoid overlapping responses.
|
| Sample derived from the second data collection round (CFA) Patients have to be aged 18 years and older, cross in an anticoagulation clinic for at least three months, have appropriate cognitive functioning as determined by the six-item screener, and have the ability to give informed consent will be the inclusion criteria. Patients will be excluded if their Charlson Comorbidity Index (CCI) is high (CCI > 4) to account for any comorbidity-related variability and assure the homogeneity of the study sample. |
Other: Participation in an web-based survey
Patients who agree to participate will access a web-based survey (SurveyMonkey) via social media, mobile messages, or emails. The invitation will provide study details, eligibility criteria, and tasks. Patients will confirm inclusion/exclusion criteria upon access, and those ineligible will opt out. Data collection will take approximately 10 minutes per patient. Collected data will be sent to an electronic data collector and saved as an eCRF file in REDCap. The study will conduct two cross-sectional data collection rounds using web surveys to ensure anonymity. Convenience sampling will target anticoagulated patient associations' webpages and social media (e.g., Feder-AIPA, AICCA). In the second round, participants will be informed not to resubmit if they have already completed the survey to avoid overlapping responses.
|
Outcome Measures
Primary Outcome Measures
- To develop and validate the first self-report measure assessing the preparedness for self-monitoring and self-management of oral anticoagulation in adults (the PERSONAE scale) [August 2023 - June 2024]
The PERSONAE scale will be developed and validated initially among Italian adult patients; however, it will be designed to potentially meet the needs for a scaling-up in diverse age groups and internationally.
Secondary Outcome Measures
- To describe the preliminary levels of preparedness for self-monitoring and self-management of oral anticoagulation in Italian adults. [July 2024 - December 2024]
The most technology-assisted learning options among the involved anticoagulated patients will be mapped to ensure future feasible technology-assisted learning initiatives based on a detailed bottom-up description of the preferences indicated by the responders. This information will provide a solid basis for designing future technology-assisted educational interventions to boost preparedness for self-monitoring and self-management of oral anticoagulation in adult patients.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
Adults (aged over 18 year old)
-
In an anticoagulation clinic for at least three months
-
Ability to give informed consent
-
CCI ≤ 4
Inclusion Criteria:
-
Age < 18 years
-
Not in an anticoagulation clinic for at least three months
-
Inability to give informed consent
-
CCI > 4
Contacts and Locations
Locations
| Site | City | State | Country | Postal Code | |
|---|---|---|---|---|---|
| 1 | IRCCS Policlinico San Donato | San Donato Milanese | Milano | Italy | 20097 |
Sponsors and Collaborators
- IRCCS Policlinico S. Donato
Investigators
- Principal Investigator: Rosario Caruso, IRCCS Policlinico S. Donato
Study Documents (Full-Text)
None provided.More Information
Publications
- Ark, L. A. van der, Koopman, L., Straat, J. H., & Bergh, D. van den. (2021). mokken: Conducts Mokken Scale Analysis (3.0.6). https://CRAN.R-project.org/package=mokken
- Callahan CM, Unverzagt FW, Hui SL, Perkins AJ, Hendrie HC. Six-item screener to identify cognitive impairment among potential subjects for clinical research. Med Care. 2002 Sep;40(9):771-81. doi: 10.1097/00005650-200209000-00007.
- Dellafiore, F., Flocco, S. F., Arrigoni, C., Barello, S., Nania, T., Russo, M. G., Sarubbi, B., Magon, A., Pittella, F., Chessa, M., & Caruso, R. (2020). Italian validation of the healthcare needs scale for youth with congenital heart disease and its short-form development. Congenital Heart Disease, 15(3), 167-180. Scopus. https://doi.org/10.32604/CHD.2020.012438
- Garcia C, Silva M, Araujo M, Henriques M, Margarido M, Vicente P, Nzwalo H, Macedo A. Admission Severity of Atrial-Fibrillation-Related Acute Ischemic Stroke in Patients under Anticoagulation Treatment: A Systematic Review and Meta-Analysis. J Clin Med. 2022 Jun 20;11(12):3563. doi: 10.3390/jcm11123563.
- Gee E, Pol A, Kittoe K, Coker F, Speed V. Keeping warfarin patients safe during the COVID-19 pandemic: review of an INR self-testing programme. Br J Nurs. 2022 Feb 10;31(3):142-146. doi: 10.12968/bjon.2022.31.3.142.
- Heneghan CJ, Garcia-Alamino JM, Spencer EA, Ward AM, Perera R, Bankhead C, Alonso-Coello P, Fitzmaurice D, Mahtani KR, Onakpoya IJ. Self-monitoring and self-management of oral anticoagulation. Cochrane Database Syst Rev. 2016 Jul 5;7(7):CD003839. doi: 10.1002/14651858.CD003839.pub3.
- Hsu JY, Liu PP, Sposato LA, Huang HK, Liu AB, Lai EC, Lin SJ, Hsieh CY, Loh CH. Oral anticoagulant decreases stroke recurrence in patients with atrial fibrillation detected after stroke. Front Cardiovasc Med. 2022 Jul 22;9:929304. doi: 10.3389/fcvm.2022.929304. eCollection 2022.
- Kantito S, Saokaew S, Yamwong S, Vathesatogkit P, Katekao W, Sritara P, Chaiyakunapruk N. Cost-effectiveness analysis of patient self-testing therapy of oral anticoagulation. J Thromb Thrombolysis. 2018 Feb;45(2):281-290. doi: 10.1007/s11239-017-1588-8.
- Kow CS, Sunter W, Bain A, Zaidi STR, Hasan SS. Management of Outpatient Warfarin Therapy amid COVID-19 Pandemic: A Practical Guide. Am J Cardiovasc Drugs. 2020 Aug;20(4):301-309. doi: 10.1007/s40256-020-00415-z.
- Kuljis J, Money AG, Perry M, Barnett J, Young T. Technology-assisted self-testing and management of oral anticoagulation therapy: a qualitative patient-focused study. Scand J Caring Sci. 2017 Sep;31(3):603-617. doi: 10.1111/scs.12374. Epub 2016 Aug 8.
- Ligtvoet, R., van der Ark, L. A., te Marvelde, J. M., & Sijtsma, K. (2010). Investigating an Invariant Item Ordering for Polytomously Scored Items. Educational and Psychological Measurement, 70(4), 578-595. https://doi.org/10.1177/0013164409355697
- Magon A, Arrigoni C, Durante A, Falchi C, Dellafiore F, Stievano A, Caruso R. Barriers to self-monitoring implementation in the oral anticoagulated population: A qualitative study. Int J Nurs Pract. 2023 Feb;29(1):e13095. doi: 10.1111/ijn.13095. Epub 2022 Aug 15.
- Magon A, Arrigoni C, Graffigna G, Barello S, Moia M, Palareti G, Caruso R. The effect of health literacy on vaccine hesitancy among Italian anticoagulated population during COVID-19 pandemic: the moderating role of health engagement. Hum Vaccin Immunother. 2021 Dec 2;17(12):5007-5012. doi: 10.1080/21645515.2021.1984123. Epub 2021 Oct 13.
- Magon A, Arrigoni C, Moia M, Mancini M, Dellafiore F, Manara DF, Caruso R. Determinants of health-related quality of life: a cross-sectional investigation in physician-managed anticoagulated patients using vitamin K antagonists. Health Qual Life Outcomes. 2020 Mar 16;18(1):73. doi: 10.1186/s12955-020-01326-y.
- Magon A, Arrigoni C, Roveda T, Grimoldi P, Dellafiore F, Moia M, Obamiro KO, Caruso R. Anticoagulation Knowledge Tool (AKT): Further evidence of validity in the Italian population. PLoS One. 2018 Aug 14;13(8):e0201476. doi: 10.1371/journal.pone.0201476. eCollection 2018. Erratum In: PLoS One. 2018 Nov 15;13(11):e0204534.
- Magon, A., Dellafiore, F., Pittella, F., & Caruso, R. (2016). Criteria and requirements of self-monitoring pathways for the management of oral anticoagulant therapy: Meta-narrative analysis of a systematic review. Ricerca e Pratica, 32(6), 246-254. Scopus.
- Marchetti G, Bertaglia E, Camerini A, De Angelis G, Filippucci L, Maggi A, Marra S, Racani C, Serrati C. Clinical Discussions in Antithrombotic Therapy Management : A Delphi Consensus Panel. J Atr Fibrillation. 2020 Feb 28;12(5):2159. doi: 10.4022/jafib.2159. eCollection 2020 Feb-Mar.
- Muthén, L. K., & Muthén, B. O. (2017). Version 7 Mplus User's Guide. Mplus. https://www.statmodel.com/download/usersguide/MplusUserGuideVer_7.pdf
- NICE. (2016). Atrial fibrillation and heart valve disease: Self-monitoring coagulation status using point-of-care coagulometers (the CoaguChek XS system) | Guidance | NICE. NICE. https://www.nice.org.uk/guidance/dg14
- Price EL, Ansell J. Virtual Education for Patient Self-Testing for Warfarin Therapy Is Effective During the COVID-19 Pandemic. Jt Comm J Qual Patient Saf. 2022 Apr;48(4):214-221. doi: 10.1016/j.jcjq.2022.01.001. Epub 2022 Jan 18.
- Rattray J, Jones MC. Essential elements of questionnaire design and development. J Clin Nurs. 2007 Feb;16(2):234-43. doi: 10.1111/j.1365-2702.2006.01573.x.
- Schicktanz S, Schweda M. Aging 4.0? Rethinking the ethical framing of technology-assisted eldercare. Hist Philos Life Sci. 2021 Aug 3;43(3):93. doi: 10.1007/s40656-021-00447-x.
- Straat, J. H., van der Ark, L. A., & Sijtsma, K. (2014). Minimum Sample Size Requirements for Mokken Scale Analysis. Educational and Psychological Measurement, 74(5), 809-822. https://doi.org/10.1177/0013164414529793
- Watson R, van der Ark LA, Lin LC, Fieo R, Deary IJ, Meijer RR. Item response theory: how Mokken scaling can be used in clinical practice. J Clin Nurs. 2012 Oct;21(19-20):2736-46. doi: 10.1111/j.1365-2702.2011.03893.x. Epub 2011 Aug 26.
- WHO. (2021). Ageing and health. https://www.who.int/news-room/fact-sheets/detail/ageing-and-health
- 52/INT/2023