Tests of calibration and goodness‐of‐fit in the survival setting

To access the calibration of a predictive model in a survival analysis setting, several authors have extended the Hosmer–Lemeshow goodness‐of‐fit test to survival data. Grønnesby and Borgan developed a test under the proportional hazards assumption, and Nam and D'Agostino developed a nonparametric test that is applicable in a more general survival setting for data with limited censoring. We analyze the performance of the two tests and show that the Grønnesby–Borgan test attains appropriate size in a variety of settings, whereas the Nam‐D'Agostino method has a higher than nominal Type 1 error when there is more than trivial censoring. Both tests are sensitive to small cell sizes. We develop a modification of the Nam‐D'Agostino test to allow for higher censoring rates. We show that this modified Nam‐D'Agostino test has appropriate control of Type 1 error and comparable power to the Grønnesby–Borgan test and is applicable to settings other than proportional hazards. We also discuss the application to small cell sizes. Copyright © 2015 John Wiley & Sons, Ltd.     

The effectiveness of interventions to reduce the household economic burden of illness and injury: a systematic review

Objective

To determine the nature, scope and effectiveness of interventions to reduce the household economic burden of illness or injury.

Methods

We systematically reviewed reports published on or before 31 January 2014 that we found in the CENTRAL, CINAHL, Econlit, Embase, MEDLINE, PreMEDLINE and PsycINFO databases. We extracted data from prospective controlled trials and assessed the risk of bias. We narratively synthesized evidence.

Findings

Nine of the 4330 studies checked met our inclusion criteria – seven had evaluated changes to existing health-insurance programmes and two had evaluated different modes of delivering information. The only interventions found to reduce out-of-pocket expenditure significantly were those that eliminated or substantially reduced co-payments for a given patient population. However, the reductions only represented marginal changes in the total expenditures of patients. We found no studies that had been effective in addressing broader household economic impacts – such as catastrophic health expenditure – in the disease populations investigated.

Conclusion

In general, interventions designed to reduce the complex household economic burden of illness and injury appear to have had little impact on household economies. We only found a few relevant studies using rigorous study designs that were conducted in defined patient populations. The studies were limited in the range of interventions tested and they evaluated only a narrow range of household economic outcomes. There is a need for method development to advance the measurement of the household economic consequences of illness and injury and facilitate the development of innovative interventions to supplement the strategies based on health insurance.     

Enhancing Stewardship of Community-Engaged Research Through Governance

Objectives. We explored the relationship of community-engaged research final approval type (tribal government, health board, or public health office (TG/HB); agency staff or advisory board; or individual or no community approval) with governance processes, productivity, and perceived outcomes.Methods. We identified 294 federally funded community-engaged research projects in 2009 from the National Institutes of Health’s Research Portfolio Online Reporting Tools, Centers for Disease Control and Prevention’s Prevention Research Centers, and Native American Research Centers for Health databases. Two hundred (68.0%) investigators completed a survey about governance processes and productivity measures; 312 partners (77.2% of 404 invited) and 138 investigators (69.0% of 200 invited) completed a survey about perceived outcomes.Results. Projects with TG/HB approval had increased likelihood of community control of resources (odds ratios [ORs] ≥ 4.80). Projects with other approvals had decreased likelihood of development or revision of institutional review board policies (ORs ≤ 0.37), having written agreements (ORs ≤ 0.17), and agreements about publishing (ORs ≤ 0.28), data use (ORs ≤ 0.17), and publishing approval (ORs ≤ 0.14).Conclusions. Community-engaged research projects with TG/HB approval had strong stewardship of project resources and agreements. Governance as stewardship protects community interests; thus, is an ethical imperative for communities, especially native communities, to adopt.Researchers working with native communities (American Indian, Alaska Native, and Native Hawaiian peoples), other racial/ethnic minority communities, or other communities facing disparities that experience similar mistrust for past research issues, health inequities (e.g., gays and lesbians or people with disabilities), or both, have advocated the use of participatory research to enhance community health.1–6 Such approaches include tribal participatory research, community-based participatory research, and participatory action research and are generally grouped as community-engaged research (CEnR). There is a continuum of engagement,7 but CEnR that involves collaborative partnership and shared leadership between community members and (academic) researchers in all phases of the research can build capacity of all partners, create research that benefits the community, and enhance translation of research findings to the community.8–13 These approaches have attraction because they can advance cocreation of the research, contribute culturally centered methods, and foster research capacity.1,2,14,15Although CEnR approaches have appeal, they still require governance to provide protection, oversight, guidance, legitimacy, and community benefit. Governance over CEnR is complex and involves numerous practices and policies.16,17 Historically, oversight responsibilities have been held by institutional review boards (IRBs) that uphold federal standards established by the Office for Human Research Protections.18,19 Use of IRBs (e.g., university IRBs or Indian Health Service IRBs) for research oversight characterizes governance as regulation as the focus is on balancing the needs of protection of individuals from harm while trying to foster scientific innovation. However, when research partners consider other functions of governance alongside legal regulation (e.g., use of tribal governments or community-based review boards), the quality of research can be strengthened and more attention paid to the benefits and harm of the research for the community.20–22In recent years, policymakers, CEnR researchers, and community organizations have advocated a broader perspective of governance, one that can be characterized as stewardship of research. Governance as stewardship enhances protection of the community, helps to foster research partnerships and appropriate access to and approval of research by community bodies, ensures benefit for the community, provides legitimacy of the research, shares responsibility for the research, provides community control, and builds research capacity in communities.20–23 For example, when native communities steward research, new patterns emerge between academic and community partners that might involve (1) community and academic partners requiring and committing to oversight by a tribal council or community board, (2) review boards or tribal governments insisting the that project demonstrate benefits to the community (not just individuals), (3) all partners committing to tribal ownership of the data, and (4) all partners working to use data and disseminate findings following tribal review.2,24–27Although nontribal communities do not have a tribal council for formal governance, they establish various governance mechanisms such as oversight by faith-based networks or leaders, health boards or public health offices, project advisory boards, or community partner boards.21,28–30 Stewardship by these governing entities may involve (1) academic partners that engage in collaboration with the community to produce the research, (2) projects that use culturally relevant research designs and instruments to enhance the quality of the research, (3) projects that hire community members on research projects to build research capacity, and (4) academic partners that encourage community engagement and participation.2–4,21,28 In both native and nonnative communities, stewardship practices lead to enhanced trust of the research process by community partners, relationships that balance community and academic institutional power, IRB processes that reflect community interests and not just biomedical interests, inclusion of cultural frameworks that fit the community, and academic members committed to community engagement.21,28,31Enhancing stewardship of research through governance has focused on several activities. First, increasingly, native and nonnative communities are asserting their roles in overseeing research by developing community IRBs and other forms of research oversight.23,32,33 Second, research review can protect community knowledge by establishing protocols for oversight and can affirm tribal or community authority to approve and guide research that will benefit the community.21,22,28–30,33,34 Third, the National Congress of American Indians35–37 asserts that tribes, as sovereign nations, have regulatory authority over research that takes place on tribal lands and with tribal citizens. Several tribes have exercised governance by establishing research codes, research review boards, and formal agreements with research institutions, and some intertribal entities have established research oversight in urban and cross-tribal regions.33,38Despite the expanded view of ethical issues within CEnR projects and an upsurge in community governance expectations from communities and some funders, there has been little research that has examined the role of governance in research specifically, as well as concerns that these processes might inhibit research. Some researchers and policy analysts suggest that tribal research review is perceived as slowing or blocking research development and dissemination.25,35 A tension related to data ownership to ensure risks and benefits are considered for communities, individual research participants, and research funders also exists.What has been lacking in these discussions to date has been research about the associations of governance with agreements, control of resources, productivity, and perceived outcomes of CEnR. Agreements are the accepted standards or protocols for the research partnership such as mission and objectives, group dynamics, and dissemination.12,39 Control of resources is whether the community, academic institution, or both hire personnel and manage project resources.12,40 Research productivity measures include garnering funding, disseminating scholarship, developing new measures centered in cultural or community perspectives, and establishing new research regulation.3,23,28,30 These measures are important as the need to generate, disseminate, and regulate new knowledge and practices are core goals of funding agencies and, to a lesser extent, communities.Perceived outcomes of CEnR focus on the contributions to health, and encompass changes in power relations, sustainability, community transformation, improved health of the community, and capacity building for individuals and agencies.12 These outcomes are important as they are health outcomes or factors that enhance public health. Ultimately, the success of a CEnR project is determined by research productivity and improvement of health outcomes.The notion of governance also has often been a source of mystery and conflict in research partnerships. We sought to foster understanding and provide context around governance as “stewardship” in research partnerships in both native and nonnative communities by focusing on the type of final approval of CEnR—the body or individual who endorsed and approved the project on behalf of the community and allowed it to continue. This approval is a key factor for legitimacy, community involvement, oversight, and guidance of the project.26,35 Furthermore, the type of approval has not been studied, whereas the general oversight of research ethics through community or tribal IRBs has garnered recent research focus.21,33,38 Examining the type of approval allows an exploration of how governance as stewardship balances needs for authority and accountability, control and capacity building, and protection and benefits.     

Impact of Preoperative Prealbumin on Outcomes After Cardiac Surgery

Background: Preoperative malnutrition is increasingly prevalent in patients undergoing cardiac surgery. Although prealbumin is a widely used indicator of nutrition status, its use in the preoperative assessment of patients undergoing cardiac surgery is not well defined. The purpose of this study is to determine the impact of preoperative prealbumin levels on outcomes after cardiac surgery. Materials and Methods: Data were prospectively gathered from February 2013 to July 2013 on 69 patients undergoing cardiac surgery. Prealbumin levels were obtained within 24 hours of surgery. Patients were divided into 2 groups based on a prealbumin cutoff value of 20 mg/dL. Results: Of the 69 patients, 32 (46.4%) had a preoperative prealbumin ≤20 mg/dL. There was no correlation between prealbumin levels and body mass index (r = ?0.13, P = .28). Likewise, there was no correlation between preoperative albumin and prealbumin levels (r = 0.09, P = .44). Nine of 32 (28.1%) patients with low preoperative prealbumin levels had postoperative infections compared with 2 of 37 (5.4%) patients with high prealbumin levels (P = .010). Patients with low prealbumin levels also had increased risk of postoperative intubation for >12 hours (P = .010). Conclusions: Patients undergoing cardiac surgery with preoperative prealbumin levels of ≤20 mg/dL have an increased risk for postoperative infections and the need for longer mechanical ventilation. If feasible, nutrition optimization of such patients may be considered prior to cardiac surgery.     

Texting-Based Reporting of Adverse Drug Reactions to Ensure Patient Safety: A Feasibility Study

Background

Paper-based adverse drug reaction (ADR) reporting has been in practice for more than 6 decades. Health professionals remain the primary source of reports, while the value of patients’ reporting is yet unclear. With the increasing popularity of using electronic gadgets in health, it is expected that the electronic transmission of reports will become the norm within a few years.

Objective

The aims of this study are to investigate whether short messaging service or texting can provide an alternative or supplemental method for ADR reporting given the increasing role of mobile phones in health care monitoring; to determine the usefulness of texting in addition to paper-based reporting of ADRs by resident physicians; and to describe the barriers to ADR reporting and estimate the cost for setting up and maintaining a texting-computer reporting system.

Methods

This was a pre-post cross-sectional study that measured the number of ADRs texted by 51 resident physicians for 12 months from the Department of Obstetrics and Gynecology and the Department of Adult Medicine of a tertiary government hospital in Manila, Philippines, with 1350-bed capacity. Reports were captured by a texting-computer reporting system. Prior to its implementation, key informant interview and focus group discussion were conducted. Baseline information and practice on the existing paper-based reporting system were culled from the records of the hospital’s Pharmacy and Therapeutics Committee. A postintervention survey questionnaire was administered at the end of 12 months.

Results

Only 3 ADRs were texted by 51 resident physicians in 12 months (reporting rate 3/51 or 6%). By contrast, 240 ADRs from the paper-based reporting system from 848 resident physicians of the study hospital were collected and tabulated (reporting rate 240/848 or 28.3%). Texting ADRs was not efficient because of power interruption, competition with the existing paper-based reporting system, and unforeseen expiration of prepaid text loads/credits. The 3 ADRs texted were a report of vivid dreams and nightmares, a report of disturbing dreams and memory lapses, both of which were due to montelukast use, and a report of hepatitis from an isoniazid/rifampicin fixed-dose combination. Nineteen of 51 resident physicians (37%) registered in the reporting system responded to the postintervention survey. The most common reasons for not reporting ADRs were no adverse reaction identified 11/19 (58%) and restrictive reporting syntax 4/19 (21%). All doctors preferred a free form of reporting. The direct cost of the texting-based reporting system was calculated to be US $5581.40 and the indirect cost was US $9989.40. The total cost for texting-based ADR reporting system for 12 months was US $15,570.79.

Conclusions

Reporting of ADRs via texting could be lower compared with an existing ADR paper-based system. Problems of Internet connectivity, reporting syntax, and expiration and reliability of text loads/credits should be addressed while implementing a text-based ADR reporting system in a developing country.