A Web-Based Application for Initial Screening of Living Kidney Donors: Development, Implementation and Evaluation

Authors


  • This research was presented in an oral presentation at the American Transplant Congress, June 2–6, 2012, Boston, MA.

Corresponding author: Derek E. Moore

derek.moore@vanderbilt.edu

Abstract

Most centers utilize phone or written surveys to screen candidates who self-refer to be living kidney donors. To increase efficiency and reduce resource utilization, we developed a web-based application to screen kidney donor candidates. The aim of this study was to evaluate the use of this web-based application. Method and time of referral were tabulated and descriptive statistics summarized demographic characteristics. Time series analyses evaluated use over time. Between January 1, 2011 and March 31, 2012, 1200 candidates self-referred to be living kidney donors at our center. Eight hundred one candidates (67%) completed the web-based survey and 399 (33%) completed a phone survey. Thirty-nine percent of donors accessed the application on nights and weekends. Postimplementation of the web-based application, there was a statistically significant increase (p < 0.001) in the number of self-referrals via the web-based application as opposed to telephone contact. Also, there was a significant increase (p = 0.025) in the total number of self-referrals post-implementation from 61 to 116 per month. An interactive web-based application is an effective strategy for the initial screening of donor candidates. The web-based application increased the ability to interface with donors, process them efficiently and ultimately increased donor self-referral at our center.

Abbreviations
BMI,

body mass index;

DM,

diabetes mellitus

HIPAA,

Health Insurance Portability and Accountability Act;

HTN,

hypertension.

Introduction

The treatment of choice for end-stage renal disease is kidney transplantation. Kidney transplantation has been well documented to increase survival and improve quality of life in comparison to conventional dialysis therapies [1-3]. Unfortunately, the number of potential recipients far exceeds the number of available donor organs and there has been minimal expansion of the deceased donor pool over the last 10 years. Living donation is a reliable means of increasing the donor supply. Despite the potential benefits of living donor kidney transplantation, the number of donors coming forward to consider donation has been relatively stable over the last ten years and live donor kidney transplants represent far less than half of all transplants performed in most centers [4].

There are multiple theories as to why living donation is not increasing even though most centers have increased efforts to encourage living donation. One potential explanation is that potential donors who lead typically busy lives are finding the task of interfacing with the transplant center too burdensome and daunting. In many programs it is likely that as many as 80% of potential donors who actually contact the center fail to donate [5-8]. This low conversion rate from self-referral to donation only adds to the already significant workload common in most living donor programs. There is little research regarding the most effective strategies to screen and efficiently identify appropriate candidates. Most centers utilize phone or written surveys to screen candidates who self-refer to be living kidney donors. However, phone screening surveys can be very resource intensive and written screening surveys are often incomplete, illegible and delay the process by several weeks when distributed through the mail. The use of technology in the process of screening has not been explored.

People in our contemporary culture are increasingly sophisticated consumers of information. They are accustomed to fast-paced, visually appealing communication media from sources such as the Internet, which offers unprecedented access to information on almost any subject [9, 10]. And with the increased awareness that social networking has raised regarding organ donation, living donor programs now have an opportunity to embrace the opportunity to use the Internet or social networking to potentially increase living donation [11-14]. Therefore, computerized interactive health communication applications on the Internet may be an ideal way to educate patients and could provide an easy, automated and cost-effective means to interface with transplant centers [13, 15]. In a system of limited health care resources, it becomes necessary to develop innovative strategies to improve cost-effectiveness of administrative and clinical practices [16].

To increase efficiency and reduce resource utilization, we developed a web-based application to screen candidates who self-refer to be kidney donors. The interactive online application for initial screening of potential living kidney donors has the potential to decrease staff resource utilization and improve processes. Web-based applications can be processed quickly, provide immediate feedback to the donor and allow staff to focus on those potential donors who are most likely to be candidates.

The aim of this study was to describe the development, implementation, and evaluation of the use of this unique and innovative web-based application. Specifically, this study evaluates: (1) the demographic characteristics of persons using the web-based application; (2) the time of usage of the web-based application; (3) the temporal course of usage of the web-based application in comparison to phone contacts; and (4) the impact of the web-based application on the numbers of persons over time who self-referred to be potential living kidney donors.

Methods

Web-based application development and description

A web-based survey was developed to provide the same information as a phone/paper screening survey, with access from our center Website as shown in Figure 1, and was implemented at our transplant center on January 1, 2011. Potential donor candidates were informed about the web-based application via written materials distributed to recipients and potential donors, on our transplant center Website and in an informational video created and distributed by our transplant center to potential donors. Individuals who called our transplant center inquiring about donation were also given the option via an automated voice recording of completing the web-based application versus speaking directly to the living donor staff for phone screening.

Figure 1.

A web-based application was developed to provide the same information as a phone/paper screening survey, with access from our center Website. The survey contains seventeen demographic and fourteen medical history questions.

This online application is easily accessible via the center Website and contains 17 demographic and 14 medical history questions. Exclusion criteria based on our center's protocols were programmed into the application. Exclusion at the point of web-based survey submission was limited to the three most common initial medical screening exclusions at our center to prevent inappropriate exclusion by the application.

The potential donor receives a real-time electronic response regarding their candidacy as a potential donor immediately after submission of the survey based on our center's inclusion/exclusion criteria. There are distinct standardized responses generated to the potential candidate based on four possible scenarios: (1) The donor is appropriate to proceed based on screening; (2) The donor is excluded because of hypertension requiring one blood pressure medication; (3) The donor is excluded because of a medical diagnosis of diabetes; or (4) The donor is excluded because of having a BMI of ≥35 kg/m2. Those candidates who are appropriate to proceed are notified via an automated response that a member of the living donor team will contact them within 48 h to proceed with evaluation. When potential candidates are excluded based on combinations of BMI ≥35 kg/m2 and/or hypertension and diabetes, the default response is a turn down based on BMI ≥35 kg/m2. This was chosen as the default because it represents a potentially correctable reason for exclusion and allows donors the opportunity to reapply. Persons who are excluded because of BMI being ≥35 kg/m2 are provided transplant center contact information in the event they would like to consider weight loss in an attempt to be a donor candidate. All automated responses contain transplant center contact information in the event the candidates would like to discuss the outcome or feel the exclusion was in error.

Submitted survey data are stored in a secure database and data encrypted to ensure HIPPA compliance. Transplant center staff can access and review the completed web-based medical survey, at which time the potential donors who passed the initial screening are contacted for additional screening.

Sources of data

This study is a retrospective cohort analysis of a prospectively collected data and retrospectively collected data at a single US transplant center. Phase I included adults who self-referred to be potential living kidney donors using the web-based application from January 1, 2011 to June 30, 2011, which was the initial implementation period of the web-based application. Phase II of the study included frequency data of all persons who self-referred for potential living donation, by either the web-based or phone systems, during the 15 months following the implementation of the web-based system (January 2011 to March 2012). The final phase of the study, phase III, included frequency data of all contacts over 27 consecutive months—the 12 months preceding the implementation of the web-based application (January 2010 to December 2010) and the first 15 months of the web-based application's use (January 2011 to March 2012). This study was reviewed and approved by the Institutional Review Board of Vanderbilt University Medical Center.

Data encoding and statistical analysis

Potential candidates for living kidney donation who contacted the transplant center from January 1, 2011 to March 31, 2012 were classified as being in either the phone or web-based application screening groups. In phase I of the study, the demographic characteristics of potential donors who self-referred using the web-based application during its initial implementation period (January 1, 2011 to June 30, 2011) were recorded: age (years), gender, race (white, African American, other) and relationship to recipient (related, including spouse or nonrelated). The time of contact was recorded as business hours or nonbusiness hours, which were classified as weeknight or weekends. Among those who were excluded from living donation using the web-based screening tool, the primary reason for exclusion was encoded as one of seven possible categories: BMI ≥ 35 kg/m2, hypertension requiring one or more medications to control, diabetes mellitus (self-report of diagnosis), BMI/hypertension, BMI/diabetes mellitus, BMI/hypertension/diabetes mellitus, diabetes mellitus/hypertension. These categories were developed to form clear, mutually exclusive reasons for exclusion for those potential donors who met multiple exclusion criteria.

Data from the initial 6-month implementation period, study phase I were characterized using frequencies and percentages. Nonparametric tests (chi-squared, Fisher's exact or comparison of medians) were used to compare the demographic characteristics of those who were and who were not excluded from consideration for donation based on their responses on the web-based application. In study phase II (January 2011 to March 2012), the proportion of contacts that were generated via phone or the web-based application over five consecutive calendar quarters were compared using Somers'd. Study phase III (January 2010 to March 2012) entailed a time series analysis over 27 consecutive months that compared the number of self-referrals per day during the pre- and postimplementation periods using the Mann–Whitney test. Statistical analyses were performed using IBM SPSS Statistics (version 20).

Results

Phase I: Implementation of the web-based application (January through June 2011)

The demographic characteristics of potential donors completing the web-based screening application during the initial 6-month implementation phase are described in Table 1. Of the 266 candidates who completed the online survey during this period, the median age was 39 years, 67% were female and 83% were white. These participants ranged in age from 17 to 73 years (Figure 2) and their age did not differ on the basis of gender (p = 0.790), race (p = 0.311), or the relationship of the potential donor to the recipient (p = 0.889). One hundred thirty-one (49%) of persons using the web-based survey were related to or a spouse of the potential recipient, 135 (51%) were unrelated and 21 (16%) of the unrelated potential donors were altruistic donors. One hundred sixty-two (61%) of the web-based surveys were submitted during normal business hours (8 am–5 pm) and 104 (39%) during non-business hours. Of those that were submitted during nonbusiness hours, 68 (65%) were week night submissions (5pm–8am) and 36 (35%) were weekend submission (Saturday–Sunday).

Table 1. Demographic characteristics of the phase I study sample
CharacteristicTotal sample (n = 266)Met inclusion criteria (n = 203)Excluded (n = 63)p-Value Included vs. Excluded
  1. Table entries are median (lower, upper 95% confidence intervals) or n (percentage).

Age (years)39 (37, 40)37 (36, 39)43 (39, 46)0.174
Gender   1.000
 Female179 (67%)137 (67%)42 (67%) 
 Male87 (33%)66 (33%)21 (33%) 
Race   0.782
 White221 (83%)169 (83%)52 (83%) 
 African American38 (14%)28 (14%)10 (16%) 
 Other7 (3%)6 (3%)1 (1%) 
Relationship to recipient   0.153
 Related (including spouse)131 (49%)105 (80.2%)26 (19.8%) 
 Not related135 (51%)98 (72.6%)37 (27.4%) 
Figure 2.

This figure depicts the wide distribution of ages of web-based application users during the first 6 months of the application's implementation.

Potential donor inclusion and exclusion by web-based application

Of those who completed the online survey between January and June 2011, 203 (76%) met inclusion criteria to proceed with initial evaluation screening for living donation and 63 (24%) were excluded based on preprogrammed exclusion criteria. Twenty-four percent of the white candidates and 26% of the African American candidates were excluded based on pre-programmed exclusion criteria. There was no difference in the race (p = 0.782), age (p = 0.174), gender (p = 1.000) or relationship to recipient (p = 0.153) of potential donors who were included versus those who were excluded using the web-based screening application.

Reasons for exclusion

A total of 63 individuals who completed the web-based application to be a possible kidney donor were excluded based on preprogrammed medical criteria. The reasons for exclusion, ordered in decreasing frequency of occurrence, are shown in Table 2. Having a BMI of ≥ 35 kg/m2 without any other reasons for screening-based exclusion was the most prevalent reason for exclusion, with 48% of potential donors being excluded for this reason. Body mass index averaged 38.8 ± 3.4 for those candidates who were excluded for this reason. The next most prevalent reason for exclusions among those completing the web-based application was hypertension requiring at least one medication (29%). Diabetes mellitus alone excluded 3% of candidates.

Table 2. Reasons for nondonation among those excluded after using the web-based application
Reason for exclusionnPercent
BMI ≥ 35 kg/m2 (autocalculated based on height and weight entry)3048
Hypertension (requiring one or more medications to control HTN)1828
BMI/HTN58
BMI/HTN/DM46
BMI/DM35
Diabetes mellitus (self-report of diabetes mellitus diagnosis)23
DM/HTN12
Total63100

Phases II: Comparison of initial web-based application and telephone contacts (January 2011 to March 2012)

Between January 1, 2011 and March 31, 2012 1200 candidates self-referred to be living kidney donors at our center. Eight hundred one candidates (67%) completed the web-based survey and 399 (33%) completed a phone survey. During the first five-calendar quarters postimplementation of the web-based application, there was a statistically significant increase (p < 0.001) in the number of self-referrals via the web-based application as opposed to telephone contact (Figure 3).

Figure 3.

There was a statistically significant increase in the use of the web-based survey compared to the telephone survey (p < 0.001) over the first 15 months of the web-based survey's availability.

Phase III: Impact of the web-based screening application (January 2010 to March 2012)

A time series analyses of the total number of self-referrals per month in the pre- and postimplementation periods demonstrated that there was a statistically significant (p = 0.025) increase in the number of living donor self-referrals after implementation of the web-based application (Figure 4). Specifically, 33% (4/12) of months before implementing the web-based survey demonstrated referral rates above the overall median and 83% (5/6) months in the initial implementation period (January–June 2011) demonstrated referral rates above the overall median. There was an overall increase in the referral rate over the 15-month postimplementation period (p = 0.025) with 73% (11/15) of the months from January 2011 to March 2012 exhibiting referral rates above the overall median. This represents an increase from an average of 61 self-referrals per month before the introduction of the web-based system to a current rate, during the first 3 months of 2012, of 116 contacts per month.

Figure 4.

This sequence chart illustrates the impact of implementing a web-based application on the number of persons being screened by our transplant center for potential living kidney donation over time. The number of contacts per month increased significantly (p = 0.025) and the most recent data show that they have almost doubled since the introduction of the web-based application.

Live donor evaluations increased from 186 in 2010 to 249 in 2011, a 25% increase in medical evaluations in 1 year. Of those 249 evaluations, 82 (33%) initiated via the web application. Live donor transplants increased from 54 in 2010 to 76 in 2011, a 29% increase in living donor transplants in 1 year. Similarly of the 76 live donor transplants performed in 2011, 17 (22%) of the donors initiated contact with the transplant center via the web application. Furthermore, between January and March 2012, 63 living donor evaluations were performed 30 (48%) of which initiated via the web application and 14 live donor transplants were performed of which eight (57%) initiated contact with the transplant center via the web application.

Discussion

Live kidney donation is a key element in efforts to close the gap between the number of wait listed candidates and available organs. In response, many programs have increased efforts to expand their living donor programs to increase living donor transplantation [17, 18]. In our program we have turned to the Internet to potentially increase donation. The Internet has established itself as a leading source of health information and interactive applications for both patients and providers [19]. However, there has been a lack of implementation, evaluation and reporting of the most appropriate forms of technology in transplantation processes and patient education.

A web-based application for the initial screening of kidney donors is an effective strategy to screen candidates who self-refer to be donors, increase access to the center and potentially increase the number of donor candidates who self-refer. In this study, we developed a web-based tool to expedite and simplify the interface between potential donors and the transplant center. After its implementation, we found that the majority of candidates who self-referred for donation used the web-based application and that there was a statistically significant increase in its usage over time compared to telephone contacts. We also found that a significant amount of use occurred after business hours and on the weekends. But perhaps the most compelling indicator of program effectiveness was that there was a statistically significant increase in the number of living donor self-referrals in the 15 months postimplementation compared to the preceding year.

One alternative explanation for the increase in donor referrals would be other simultaneous efforts to recruit new donors during the period of web-based application implementation. At this time we cannot account for external confounding factors such as potential economic influences, but there were not additional internal efforts to recruit new donors during the study period.

Another alternative explanation for the increase in donor referrals would be a corresponding increase in recipient referrals. However, at our center, the number of recipient referrals has remained stable in the year before implementation (932 in 2010) and the subsequent periods post implementation (936 in 2011 and 203 in January–March 2012). This demonstrates that increased living donor referrals are not a response to increased recipient referrals. Similarly, we found that live donor medical evaluations increased by 25% and live donor transplants increased by 29% in one year. Of the evaluations and live donor transplants performed in 2011, 33% of the donor evaluations and 22% of the live donor transplants initiated contact with the transplant center via the web application. Typically, our donor evaluation and surgery scheduling takes 3–6 months; therefore, it was not until 6 months after implementation of the application that the number of donor evaluations and live donor transplants began to be impacted. The data between January and March 2012 shows that 48% of donor evaluations and 57% of transplants originated from the web application.

The web-based application provides recipients with a convenient method to enable family members and friends to initiate contact with the transplant center and to be screened. Another potentially unforeseen benefit of the web-based application is that it allows potential donors to keep their interest in donation private. They can choose to check their eligibility before disclosing their interest in donation. Potential donors can then disclose their interest and eligibility for donation when they feel comfortable with the decision to donate without perceived pressure from recipients or coordinators over the phone.

There is a lack of data specifically describing the substantial resource utilization and time involved in operating a living kidney donor program. In one of the few studies examining living donor program resource utilization, Saunders et al. concluded that the workload is significant and the donor evaluation is time-consuming in common living donor programs [20]. The online format makes it possible to screen large numbers of candidates quickly. At our center, we found that the web-based application saved our coordinators approximately 10–12 min per donor contact. For the 801 candidates screened with the web-based survey, this equates to a potential time savings of almost 160 h. The time saved by eliminating the need for telephone interviews has been replaced by the time coordinators spend working up the larger numbers of “eligible” donors. The web-based application has increased referral volume and placed new demands on the coordinator's time. However, the preprogrammed exclusion criteria, which excluded 24% of ineligible candidates before coordinator contact, allows staff to focus on those potential donors who are most likely to be candidates. The application is a low cost approach to screen out ineligible candidates, decreasing the time nurse coordinators spend on candidates who will not translate into potential donors. Aside from the initial design, internal programming, and piloting, the web-application requires minimal ongoing maintenance. If the web-based application results in only one additional live donor transplant, the cost of implementing this tool is more than accounted for.

This study did not show a concentration of use of a web-based tool among persons of any specific age group and users ranged in age from 17 to 73. However, it is likely that disparities may exist and future work will explore association of race, age, ethnicity, socioeconomic status and health literacy in greater detail. It is important to understand who is using this web-based tool and, more importantly, to understand the characteristics of persons who are either not using it or who do not have access to the tool [21-24].

We found that 24% of candidates who completed the web-based screening were excluded based on preprogrammed criteria. Alarmingly, 67% of ineligible donors were excluded based on elevated BMI or a combination of BMI, DM and HTN. This speaks to the rising rate of obesity in the United States and especially in the Southeast [25-27]. The importance of this finding is that it highlights a significant proportion of individuals who are willing to consider donation but are unsuitable because of a potentially modifiable condition. This presents an opportunity to provide the donors with education and motivation to lose weight to not only improve their health but also the opportunity to donate and improve the recipients’ health as well.

There was no difference in candidates’ inclusion or exclusion by web-based screening based on race, age, gender and relationship to the recipient. It has been widely reported that minorities are underrepresented in the donor pool [18, 28-32]. However, based on our initial analysis, it does not appear that African Americans who are screened via the web-based application are being excluded more often than white candidates because of body mass index, hypertension or diabetes. Although minorities are not being excluded at a different rate after they contact the transplant center, it is likely that a substantial proportion of minorities are appropriate candidates but fail to initiate contact [8, 18, 33]. One reported explanation is that minorities fail to even initiate or complete the donor process because they have preconceived fears or attitudes toward donation [32-37]. The web-based tool could be used to combine educational information or support tools with the survey to help allay any fears or misconceptions that prevent donation [13, 38].

One of the limitations of this study is that it was designed to assess the impact of the web-based application on donation and unfortunately a direct comparison of web-based applicants and telephone applicants was not performed. This will be analyzed in future studies. Another limitation of this study is that nonwhite, non-African American candidates comprise a very small proportion (in our study less than 3%) of the persons who completed the web-based survey. This is consistent with the racial population of recipients listed at our transplant program. Regardless, it is impossible to evaluate the use of this tool in nonwhite, non-African American candidates because of small sample size. It could be possible that this group of candidates could have initiated the donation process by phone as opposed to the web-based application. Future research to compare web-based survey candidates with phone survey candidates in terms of demographic characteristics is underway. Another potential limitation is the lack of user-reported data concerning satisfaction and ease of use of the web-based tool. Future revisions will include questions addressing satisfaction and ease of use so that we can revise our application based on user input [24].

In summary, the web-based application has enhanced our ability to attract potential donor candidates, interface with them and process them more efficiently. We have demonstrated that its use has resulted in a substantial increase in the number of potential donors coming forward to be screened. Thus far, our research demonstrates that a web-based screening application can be successfully developed and implemented in a living donor program. Perhaps one of the most important benefits of this new tool is the ease with which it can be incorporated into many other transplant programs. Further research is needed to determine if the findings after the implementation of the web-based application can be duplicated at other transplant centers. The ultimate goal for the web-application would be to have future iterations of the application distributed to multiple centers and posted on as many organ donation and transplantation education Websites as possible. By making information and access to living kidney donation easy and widely available, we believe living kidney donation can be increased.

Acknowledgments

Authorship statement: I, Deonna Moore, had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Ancillary