Acceptability of pictographs as a novel patient identifier to improve patient safety in the neonatal intensive care unit

Abstract

As part of a randomized controlled trial on the use of pictographs (images used in lieu of a patient photo) embedded in the electronic health record to reduce wrong-patient errors in the neonatal intensive care unit (NICU), we conducted a series of surveys of parents, providers and nurses in the NICU. Data from survey responses were thematically analyzed and categorized. We found that in all groups, there was very high awareness of the intended purpose of the pictographs; however, the perception of providers and nurses about the effectiveness of pictographs was not as strong. While several providers and nurses acknowledged that pictographs can or have helped them avoid wrong-patient errors when caring for multiple birth infants (such as twins), many nurses believed that their current practice of the use of two patient identifiers was sufficient, and pictographs were not useful. Parents reported that pictographs improved their experience of care.

Introduction

Wrong-patient errors are a serious threat to newborn safety, and newborns are shown to be at a significantly higher risk for this type of error compared to adults and other children.^1-2 Serious harm is caused when these errors reach the patient, e.g., when a medication is administered to the wrong infant.^3-4 According to the Joint Commission, this risk is in part due to the use of temporary names for infants.⁴ While the Joint Commission has required the use of more distinctive temporary names that include a portion of the mother’s name (such as “Wendy’s Baby Boy”)^5-6 wrong-patient error persists in the neonatal intensive care units (NICUs) where multiple birth infants, whose names potentially differ by one digit or letter (i.e. Baby Girl A or B) and who share a date of birth, are often concurrently admitted.⁷ Naming errors among multiple birth infants are also reported, indicating that solutions focused on naming alone cannot fully mitigate this type of error.⁸ Moreover, wrong-patient errors are also a well-recognized problem in the adult setting, where patient names are usually distinctive.⁹ Collectively, this evidence indicates that alternative methods of patient identification are required to reduce wrong-patient errors and improve patient safety.

In the adult setting, it has been shown that displaying the patient’s photograph in the electronic health record (EHR) system result in a substantial reduction in wrong-patient errors.¹⁰ However, given the prevalence of concurrent admission of multiple birth infants (such as twins, triplets, and higher order multiples) in the NICU, and considering that neonates have underdeveloped distinguishing facial features, particularly when born preterm, patient photographs do not seem to be an appropriate solution in this population. We have proposed the use of pictographs instead; a pictograph is an image of an object (such as a toy, a type of food, etc.) that is appealing and distinctive. In a multicenter randomized controlled trial (RCT) funded by the Eunice Kennedy Shriver National Institute for Child Health and Human Development (5R01HD094793)¹¹ we are studying the efficacy and safety of pictographs for reducing wrong-patient errors in the NICU. In this trial, parents/guardians select a unique pictograph for each infant admitted to the NICU, and the pictograph is both uploaded into the EHR and displayed in lieu of a patient photo, as well as printed and affixed near the infant’s hospital bed. There are more than 420 pictographs to select from, and they were designed to be visually distinctive, easily recognizable, and free from sensitive connotations (e.g., religious connotations). Additionally, a smaller version of the pictograph is printed and placed near the entrance to the patient room. The pictograph is intended to be a visual cue in the EHR to help prevent patient misidentification; it is not considered an identifier. The RCT is ongoing and quantitative results are being collected. In this manuscript, we report preliminary results of a qualitative evaluation we conducted about the acceptability of pictographs and their perceived usefulness by parents, providers, and nurses in the NICU.

Methods

This study was conducted in the NICU of Brigham and Women’s Hospital, Boston, MA, which is a large teaching hospital that is part of the Mass General Brigham health system. This level III NICU has a total of 66 beds, a majority of which are occupied by patients in need of continuous intensive care for several weeks or more. We conducted three electronic surveys, one of parents/guardians of infants in the NICU, one of the NICU ordering providers, and one of the NICU nurses. The surveys were anonymous (i.e., they did not collect any information about the respondent). They were created in REDCap¹² and a link to the survey was emailed to select providers most closely involved with placing orders in the NICU (attending physicians, fellows, nurse practitioners and physician assistants) and clinical nurses who worked at this NICU. For parents, the survey was loaded on a tablet device which was handed to the parents to fill out; alternatively, the survey link was sent to the parents in an email so they could fill it out later if desired. The questions for the surveys were developed by a group of researchers at Brigham and Women’s Hospital and were reviewed and validated by research collaborators from the other study sites. The study was approved by the Institutional Review Board at Mass General Brigham.

The parent survey contained questions about how the parents first learned about the pictographs, whether they agreed to select a pictograph for their infant(s), what their impression was of the pictographs and the selection process, what they liked or did not like about the pictographs, whether and how they found pictographs to be useful, whether they would recommend that other parents use the pictograph and why, and their suggestions about how to improve the project as a whole. The nurse and provider surveys used identical questions which concerned their understanding of why pictographs are being used, how they interacted with pictographs as they provided care, their perceptions on the usefulness of pictographs (in general, and specifically for patient identification), facilitators and barriers to the use of pictographs, other ways in which pictographs could be used, and their suggestions about how to improve the project as a whole. Box 1 shows the full text of questions included in each survey.

Box 1 -. Questions included in each of the surveys used in this study.

Parent Survey

1. Please tell us about how you first learned of baby pictographs. What were your initial impressions?

2. Did you agree to display a pictograph for your baby (babies)? Why or why not?

3. Please tell us about the pictograph selection process. How were you involved in selecting the pictograph? How did this involvement feel?

4. What were your impressions of the pictograph images? Did the images seem appropriate? Were there other images that you would have liked to see?

5. What do you like about the baby pictographs?

6. What do you NOT like about the baby pictographs?

7. How would you describe the usefulness of these pictographs?

8. What aspects of the baby pictograph program could be done better?

9. What advice or information would you give to other parents with babies in the NICU about this program?

Provider/Nurse Survey

1. What is the rationale for the use of baby pictographs?

2. How do you interact with these pictographs as you provide patient care?

3. How would you describe the usefulness of these pictographs?

4. What are your thoughts about how pictographs could assist you in identifying infants in the NICU?

5. How do pictographs help you to discern infants from each other? For example, could you picture who is “the Baseball” and who is “the Rainbow”?

6. What poses barriers to pictograph use?

7. What facilitates pictograph use?

8. In what other ways could pictographs be used?

9. What aspects of the baby pictograph program could be improved?

For questions concerning the usefulness of pictographs, one researcher categorized the responses into three groups of highly useful, sometimes or minimally useful, and not useful. We report the count and percentage of responses in each category. Given the descriptive nature of this qualitative study, convenience sampling was used, and statistical analyses were limited to direct comparisons of responses from different groups of surveyed individuals. These comparisons were done using the chi-squared test or Fisher’s exact test, based on the distribution of the responses to each question.

Results

The survey was responded to by parents of 7 infants, and by 22 ordering providers and 43 nurses. Not all surveys responses were complete (i.e., included an answer to every question). All questions were answered in 6/7 (86%) of the parents’ surveys, 20/22 (91%) of the providers’ surveys, and 34/43 (79%) of the nurses’ surveys. We included all data in the analysis; therefore, the denominators for categorical analyses were different for some questions compared to others.

Results of the Parent Surveys

All responding parents had selected a pictograph for their infant(s). They all shared a positive general impression about the project and about the pictographs. They described it as “fun yet benign”, “worthwhile”, “taking no effort”, “interesting”, and one parent simply wrote “love it.” In their response to whether they agreed to select a pictograph, most provided an additional statement justifying their decision (even though the question did not specifically ask for it), and most cited the potential value of pictographs in patient identification as a reason. This was often in their own words, such as “made our baby feel unique” or “my son was a 24 weeker and his physical [features] were really small”.

When asked about the process of selecting a pictograph, all parents reflected on their involvement in the selection process, and 2/6 (33%) indicated that having the choice (as opposed to being assigned a pictograph) was a positive experience. Also, 3/6 (50%) of parents indicated that they made their choice based on what the pictograph meant to them or based on a relationship between the pictograph and the name they had selected for their child. One parent indicated that they would have loved to have been given a longer time to choose the pictograph among the available options, citing that “it would have been a positive thing to focus on during such a scary and traumatic time.” Two parents indicated the large number of choices to select from was helpful, while a different parent indicated that they wished to see the other options that were not available to them (i.e., those which were in use for other infants).

All (100%) of the parents indicated that the pictograph choices were appropriate, and they described them with keywords such as “playful”, “fun”, “cute” and “easily recognizable.” Two parents stated that using the pictograph “added personality to the babies”, and another reported that “it provides something nursery-like for parents that don’t have their babies in a nursery at home.” To the question asking what they did not like about pictographs, all (100%) parents responded with “nothing”. Parents also mentioned other utilities to pictographs beyond their potential value in patient safety for providers: 3/6 (50%) of parents indicated that pictographs helped them when they logged into the patient portal to verify which of their children’s records they were viewing.

When asked how the use of pictographs could be improved, 4/6 (67%) of parents had no suggestions, one parent suggested allowing more time for pictograph selection, and one parent recommended assigning the pictograph as early as possible (“right from admission”). When asked if they had any suggestions for other parents about the use of pictographs, 3/6 (50%) of parents explicitly stated they would recommend it, and the remaining parents emphasized the things they valued about pictographs such as being easy to select, harmless, and helpful.

Results of the Provider and Nurse Surveys

All 22 providers (100%) and 40/43 nurses (93%) correctly responded to the question about the rationale for using pictographs (p-value 0.5452 for comparison between providers and nurses). The other 3 nurses stated that they were unsure of the point of using pictographs.

When asked about how they interacted with the pictographs, 12/22 (55%) of providers mentioned that they used them to make a connection between the pictograph at the bedside and the one in patient’s chart, while the rest either said they minimally interacted with the pictographs or did not interact with them at all. It is worth noting that some attending providers rarely place orders and have few direct interactions with the patient’s chart, as most of the orders are placed by the physicians in training or advance practice providers they oversee. While one provider stated that the pictographs have no meaningful connection to the baby, two other providers stated that they sometimes learned from the families about the reason the pictograph was selected, and this helped them associate the pictograph with the infant and the infant’s name. The fraction of nurses who mentioned they associated the pictographs with the patients was similar (25/43 or 58%, p-value 0.797). Unlike providers, none of the nurses included any comments about the meaningfulness of pictographs in their responses to this question.

Of 21 providers who responded to the question about usefulness of pictographs, 4 (19%) described them as highly useful—2 of whom particularly emphasized its usefulness in distinguishing multiples—and 13 (62%) described the pictographs as somewhat or minimally useful, while 4 providers (19%) described pictographs as not useful at all. In comparison, of 39 nurses who responded to this question, 12 (31%) described them as highly useful—with 3 calling out their value in distinguishing multiples—8 (21%) described them as somewhat or minimally useful—of whom 2 described the usefulness only to the parents (e.g., “families seem to like and enjoy picking them”)—and 19 (49%) described them as not useful at all (chi-squared p-value 0.005 for difference between the categories for providers and nurses). Among the nurses who described pictographs as not useful, one indicated that because they only have a few patients under their care and see the patient regularly, the traditional identifiers (such as patient’s name and medical record number) are enough, and pictographs don’t have any additional value for them. Another nurse commented that while the pictographs were not useful to her, she could understand why they might be useful to providers.

When asked specifically about the potential value of pictographs in improving patient identification, 8/20 (40%) of providers provided positive feedback; this included comments regarding helpfulness in distinguishing twins, in identifying long-term patients, or that pictographs are easier to remember than names. Several of the providers who did not describe the pictographs as useful in identification of patients had noted that their usefulness could have been improved if they were placed in more prominent locations, if they were printed in larger size, and if they were also displayed on printed documents such as patient handoff lists. Notably, some providers conduct some or most of their rounds outside the patient room, and the small size of the printed pictograph that was placed near the patient room entrance may have contributed to their perception of the low usefulness of pictographs. In comparison, 15/38 (39%, p-value 0.969) of nurses described pictographs as useful in identifying patients, 4 of whom mentioned their value in distinguishing multiples. Three nurses indicated that pictographs could have been useful if they also appeared in other places such as on the patient’s wristband, on the crib card (a card attached to the patient’s crib which also contains their identifiers), or on the breastmilk bottles stored in the refrigerator. Two of the nurses said that because they always use two identifiers to identify patients, they don’t think this additional identifier is necessary or helpful.

Only 17 (77%) providers responded to the question about barriers to using pictographs. Some responses were not categorized into a theme (e.g., “lack of utility in my role”); the emerging themes among the other responses included issues around the size and placement of printed pictographs, and the fact that providers rarely hear the story behind why the parents selected a specific pictograph and cannot make a memorable connection between the pictograph and the patient as a result. Only 27 (63%) of nurses responded to this question, and we found additional themes in their responses. Some nurses stated that because pictographs get reused for different patients over time, this could result in confusion or errors. Other nurses emphasized that because patients are moved from one bed to another during the NICU stay, failure to consistently move the pictographs along with the infant may be a barrier to the effective and safe use of pictographs. One nurse stated that “some [pictographs] are offensive to some cultures” but did not specify an example, and one nurse stated that because pictographs are not a legally acceptable patient identifier, some nurses may not use them and instead focus on acceptable identifiers such as patient’s name and medical record number.

When asked about other ways pictographs could be useful, only 7 (32%) of ordering providers gave a response, which included “using them to decorate the patient room”, “placing them on other printed documents [about the patient]”, and by including them in an alert for patient verification. Of the 19 (44%) of nurses who responded to this question, 4 mentioned using them for labeling breastmilk bottles, 2 mentioned including them on medication labels, 2 mentioned the value of pictographs in starting conversations with parents about their choice and the meaning of the pictograph to them, and other ideas proposed included using related pictographs for patients that are in each area of the NICU, creating memorabilia (e.g., a “coloring book”) for the patient and family based on their selected pictograph, and allowing for selection of pictograph prenatally. Six providers and 2 nurses mentioned that the pictographs could have been more effective if they were printed larger, and one provider mentioned that they could be more effective if assigned early on the day of admission. Two of the nurses stated that they were uninformed about the purpose of the pictographs (e.g., “at first, it was difficult to understand why there were random [pictographs] everywhere”), and two stated that expanding the list of pictographs to contain more items would have been helpful.

Discussion and Conclusions

We found that nearly all surveyed parents, providers and nurses could explain the intended purpose of pictographs, and parents had generally positive sentiments about the pictographs and the process of selecting them. Some parents emphasized the significance of the pictographs to them, and some providers and nurses highlighted that knowing this significance helped them in associating the pictograph with the patient. In all surveyed groups, there were responses emphasizing that enabling the selection of pictographs sooner during the admission (or even prenatally) might further increase their (presumed) beneficial effect on patient safety. Among providers and nurses, respondents had acknowledged the value of pictographs in better associating patient charts with the individual patients, particularly when accessing records associated with multiple birth infants (such as twins). Notably, some parents also shared a similar value of pictographs for them, when accessing the patient portal for their multiple children.

While all providers and nearly all nurses were able to convey the intended purpose of pictographs (to improve patient identification), their belief about the pictograph’s ability to achieve this goal was variable. More providers than nurses believed that pictographs were useful to them. This is expected, as the pictographs are primarily designed to help providers avoid wrong-patient order errors. Also, the nurses reported having fewer patients under their care at any shift, typically a maximum of two to three patients per nurse per shift, (i.e., knowing their patients well) and regularly relying on the of use of two identifiers for identifying patients before administering medications, drawing laboratory samples, etc. The use of two identifiers in this manner is a requirement by the Joint Commission.¹³ Together, these may explain why the perception of nurses about the value of pictographs was different than the perception of providers. However, it is worth noting that findings from prior research which indicates a significantly higher rate of wrong-patient errors among infants compared to general pediatric patients, and an even higher rate among multiple birth infants.⁷

Another factor may have contributed to why some nurses believed pictographs were not useful to them. In this study site, the process of enrollment of infants into the study and selection of pictographs by their parents was assisted by a group of research nurses (not the bedside nurses), which is the standard of practice at the study site. Therefore, bedside nurses were not significantly involved in the project, and despite the communications shared during project initiation, they may not have been as engaged and informed in the project as providers and parents. It is possible that if bedside nurses were more engaged in the process and more aware of the significance of pictographs to the parents, they would have a different perception of the usefulness of pictographs as well. In the multi-center RCT, we are using different enrollment processes at different sites, and another site is specifically using the help of bedside nurses to introduce the study to parents and assist them in the process of pictograph selection. Our future research includes a similar survey study at the other study sites, which will potentially demonstrate if these different implementation strategies are associated with different perceptions of pictograph usefulness among nurses and other groups.

Some of the comments from the nurses and providers are notable in that they speak to elements of the RCT that were not specifically explained to them. For instance, some nurses had indicated that the reuse of pictographs for different patients over time may itself cause a risk for patient identification errors; in this RCT, pictographs are removed from the patient’s chart shortly after the patient is discharged from the NICU, and they remain inaccessible for a 7-day “washout period”, after which they would be available for selection by parents of other infants. It is hypothesized that this washout period will provide a safeguard against the type of error raised by those survey respondents. One nurse had mentioned that the pictographs could be offensive for some cultures; as part of the RCT, all pictographs were assessed by a panel of reviewers, and those which were deemed as potentially sensitive were removed, including all pictographs with a known religious connotation (e.g., cross, six-pointed star, or crescent of the moon). Moreover, parents had full authority on the selection of pictographs for their infants, which could help not using a pictograph for a child in a way that could have a negative connotation (such as using the firetruck icon for a child who had lost a parent in fire). Finally, nurses mentioned that if pictographs are not moved with the patient when the patient is being transferred to a different bed within the same NICU, or their transfer is delayed, this could impact their usefulness; this was a known risk to the study, and the printed pictographs contained a footnote that reminded the staff to move them when they transfer the patients; additionally, the study staff were regularly reviewing the patient list and rounding the NICU to identify any pictographs that were left behind and moved them to the new location of the patient as soon as possible.

One of the providers stated that incorporating the pictographs into an alert may improve their effectiveness. While this may be true, alerting the EHR users for the sake of patient verification can result in significant burden on the users and lead to alert fatigue. Indeed, a previous study on the use of patient verification alerts in the adult setting to reduce wrong-patient errors¹⁴ was criticized for the burden of these alerts on the users.¹⁵ The premise behind the use of patient photos to reduce wrong-patient errors in adult patients is that they provide a non-interruptive solution that is as effective as patient verification alerts.¹⁰ The RCT will provide evidence as to whether pictographs have a similar effect in the neonatal care setting.

This study has several limitations. The survey responses are from a small group of conveniently sampled parents, providers and nurses and may not be generalizable. The survey can be subject to self-selection bias (e.g., individuals with more negative perspective may be more likely to respond), and as an anonymous electronic survey, it is possible that the same person provided multiple survey responses, although we did not see any evidence for it. The survey instrument was not extensively validated, and findings from this preliminary study will inform the subsequent larger scale survey we will be conducting at the other study sites. There was no formal sample size calculation and the sampling method (convenience sampling) and inherent limitations of an anonymous electronic survey may impact the generalizability of our results. Nonetheless, the results of this study are directionally consistent between the surveyed groups, and they align with the findings from prior research about the significance of wrong-patient errors among multiple birth infants.

In conclusion, our qualitative study of the perceptions of parents, providers and nurses about the use of pictographs as a means to potentially reduce wrong-patient errors showed that all of these groups had an excellent understanding of the premise for using pictographs, and many of them believed that pictographs were able to achieve this goal. Quantitative results from the ongoing RCT on the effectiveness of pictographs in reducing wrong-patient errors will help determine how much these perceptions correlate with objective results of wrong-patient errors in the NICU setting when using pictographs. Nurses, in particular, were more critical of the value of pictographs for patient identification. Parents, on the other hand, found additional sentimental value in the selection of pictographs.