Missing data approaches in eHealth research: simulation study and a tutorial for nonmathematically inclined researchers

Background

Missing data is a common nuisance in eHealth research: it is hard to prevent and may invalidate research findings.

Objective

In this paper several statistical approaches to data “missingness” are discussed and tested in a simulation study. Basic approaches (complete case analysis, mean imputation, and last observation carried forward) and advanced methods (expectation maximization, regression imputation, and multiple imputation) are included in this analysis, and strengths and weaknesses are discussed.

Methods

The dataset used for the simulation was obtained from a prospective cohort study following participants in an online self-help program for problem drinkers. It contained 124 nonnormally distributed endpoints, that is, daily alcohol consumption counts of the study respondents. Missingness at random (MAR) was induced in a selected variable for 50% of the cases. Validity, reliability, and coverage of the estimates obtained using the different imputation methods were calculated by performing a bootstrapping simulation study.

Results

In the performed simulation study, the use of multiple imputation techniques led to accurate results. Differences were found between the 4 tested multiple imputation programs: NORM, MICE, Amelia II, and SPSS MI. Among the tested approaches, Amelia II outperformed the others, led to the smallest deviation from the reference value (Cohen’s d = 0.06), and had the largest coverage percentage of the reference confidence interval (96%).

Conclusions

The use of multiple imputation improves the validity of the results when analyzing datasets with missing observations. Some of the often-used approaches (LOCF, complete cases analysis) did not perform well, and, hence, we recommend not using these. Accumulating support for the analysis of multiple imputed datasets is seen in more recent versions of some of the widely used statistical software programs making the use of multiple imputation more readily available to less mathematically inclined researchers.     

Using the Consolidated Framework for Implementation Research to Identify Barriers and Facilitators for the Implementation of an Internet-Based Patient-Provider Communication Service in Five Settings: A Qualitative Study

Background

Although there is growing evidence of the positive effects of Internet-based patient-provider communication (IPPC) services for both patients and health care providers, their implementation into clinical practice continues to be a challenge.

Objective

The 3 aims of this study were to (1) identify and compare barriers and facilitators influencing the implementation of an IPPC service in 5 hospital units using the Consolidated Framework for Implementation Research (CFIR), (2) assess the ability of the different constructs of CFIR to distinguish between high and low implementation success, and (3) compare our findings with those from other studies that used the CFIR to discriminate between high and low implementation success.

Methods

This study was based on individual interviews with 10 nurses, 6 physicians, and 1 nutritionist who had used the IPPC to answer messages from patients.

Results

Of the 36 CFIR constructs, 28 were addressed in the interviews, of which 12 distinguished between high and low implementation units. Most of the distinguishing constructs were related to the inner setting domain of CFIR, indicating that institutional factors were particularly important for successful implementation. Health care providers’ beliefs in the intervention as useful for themselves and their patients as well as the implementation process itself were also important. A comparison of constructs across ours and 2 other studies that also used the CFIR to discriminate between high and low implementation success showed that 24 CFIR constructs distinguished between high and low implementation units in at least 1 study; 11 constructs distinguished in 2 studies. However, only 2 constructs (patient need and resources and available resources) distinguished consistently between high and low implementation units in all 3 studies.

Conclusions

The CFIR is a helpful framework for illuminating barriers and facilitators influencing IPPC implementation. However, CFIR’s strength of being broad and comprehensive also limits its usefulness as an implementation framework because it does not discriminate between the relative importance of its many constructs for implementation success. This is the first study to identify which CFIR constructs are the most promising to distinguish between high and low implementation success across settings and interventions. Findings from this study can contribute to the refinement of CFIR toward a more succinct and parsimonious framework for planning and evaluation of the implementation of clinical interventions.

ClinicalTrial

Clinicaltrials.gov NCT00971139; http://clinicaltrial.gov/ct2/show/NCT00971139 (Archived by WebCite at http://www.webcitation.org/6cWeqN1uY)     

Moving research into practice: A decision framework for integrating home telehealth into chronic illness care

Home telehealth is an effective alternative for some aspects of traditional care in chronic illnesses such as diabetes, congestive heart failure and asthma. However, in spite of evidence to support use of home telehealth technologies, they have not been adopted as predicted. A significant challenge for decision-makers is applying results from multiple small studies to the care of large numbers of clients in a health region. Aside from the technology, this complex decision must also include expected client outcomes, variations in nursing resources and their deployment in service delivery. This paper presents research evidence supporting the effectiveness of home telehealth for diabetes care, with attention to the range of technologies and outcome measures reported. It also discusses implications of a recently released national study on "Homecare Indicators" that reported resource allocation and outcomes in home care. The burden of illness, evidence of technology effectiveness and proposed home care outcome indicators are considered together in a decision framework to demonstrate an approach for decision-makers and practitioners to transfer home telehealth research into practice. The resulting decision framework is applied to diabetes care within one large health region in Canada to illustrate its utility as a research transfer strategy.     

Internet-based patient education and support interventions: a review of evaluation studies and directions for future research

The Internet provides patients, families, and health providers with unparalleled opportunities to learn, inform, and communicate with one another. This paper is a review of studies that have evaluated the impact of professionally facilitated Internet-based programs for diverse clinical populations on health outcomes, utilization, and user satisfaction. The findings suggest that some outcomes in certain groups can be moderately improved and that user satisfaction has been generally positive. Given the increasing need to provide timely and effective patient-centered care, a stronger evidence base for eHealth applications must be established before broader deployment of such programs will take place.     

eHealth Program to Empower Patients in Returning to Normal Activities and Work After Gynecological Surgery: Intervention Mapping as a Useful Method for Development

Background

Full recovery after gynecological surgery takes much longer than expected regardless of surgical technique or the level of invasiveness. After discharge, detailed convalescence recommendations are not provided to patients typically, and postoperative care is fragmented, poorly coordinated, and given only on demand. For patients, this contributes to irrational beliefs and avoidance of resumption of activities and can result in a prolonged sick leave.

Objective

To develop an eHealth intervention that empowers gynecological patients during the perioperative period to obtain timely return to work (RTW) and prevent work disability.

Methods

The intervention mapping (IM) protocol was used to develop the eHealth intervention. A literature search about behavioral and environmental conditions of prolonged sick leave and delayed RTW in patients was performed. Patients’ needs, attitudes, and beliefs regarding postoperative recovery and resumption of work were identified through focus group discussions. Additionally, a literature search was performed to obtain determinants, methods, and strategies for the development of a suitable interactive eHealth intervention to empower patients to return to normal activities after gynecological surgery, including work. Finally, the eHealth intervention was evaluated by focus group participants, medical doctors, and eHealth specialists through questionnaires.

Results

Twenty-one patients participated in the focus group discussions. Sufficient, uniform, and tailored information regarding surgical procedures, complications, and resumption of activities and work were considered most essential. Knowing who to contact in case of mental or physical complaints, and counseling and tools for work reintegration were also considered important. Finally, opportunities to exchange experiences with other patients were a major issue. Considering the determinants of the Attitude–Social influence–self-Efficacy (ASE) model, various strategies based on a combination of theory and evidence were used, resulting in an eHealth intervention with different interactive functionalities including tailored convalescence recommendations and a video to communicate the most common pitfalls during the perioperative period to patients and employers. Fifteen patients in the focus groups, 11 physicians, and 3 eHealth specialists suggested points for improvement to optimize the usability of the eHealth intervention and judged it an approachable, appropriate, and attractive eHealth intervention to empower gynecological patients.

Conclusions

The IM protocol was a useful method to develop an eHealth intervention based on both theory and evidence. All patients and stakeholders judged the eHealth intervention to be a promising tool to empower gynecological patients during the perioperative period and to help them to return to normal activities and work.