Family presence during invasive procedures: a pilot study to test a tool

The Twibell et al. [31] questionnaire was selected and translated into Spanish, where the opinions, experiences, and perceptions of the medical and nursing staff about resuscitation witnessed by the family were explored [32]. This tool was modified to probe the same parameters, but in relation to invasive procedures (IP). In turn, due to the degree of invasiveness of the procedures, they were divided into high, medium or low complexity according to the Guidelines for Performance of Invasive Procedures by Medical Students of Yale University School of Medicine [33]. The new tool consisted of three questionnaires related to FPDIP, a socio-demographic one, a Risk-Benefit assessment questionnaire (FPRB) and a Self-Confidence evaluation questionnaire (FPSC). The FPRB scale consisted of 22 items with 5 points on the Likert scale and ranged from strongly disagree (1) to strongly agree (5). A high score indicated a high level of perceived benefit from the FPDIP. The FPSC consisted of 15 items on self-confidence, from not being confident (1) to very confident (5). A high score indicated a high level of self-confidence to handle the FPDIP. Subsequently, a previous pilot test was carried out to evaluate the understanding of the questionnaire questions by 25 members of the health personnel, and a Delphi group made up of experts in different subjects was convened and a final draft was reached. Finally, an item with understanding difficulties was identified; item 9 was changed and items 29 and 30 were removed because they did not proceed in the IP (See Table 1).

The 541 professionals who are part of the medical and nursing staff of the Consorci Sanitari del Garraf (CSG) were invited via corporate email to participate. There were 344 nurses and 197 physicians.

In the email, they were informed of the characteristics of the study and sent the link to Google forms so that they could answer online. The survey was also distributed in paper format by the different services of the CSG, and a box was provided for the collection of completed questionnaires. The data collection period was from September 1, 2019, to November 30, 2019.

To measure the performance of the instrument used, a correlation analysis of the items and the scale as a whole (item-test correlation) was performed. The retention criterion for each item was that the correlation was significant, following the indications of Crocker and Algina [34]. Cronbach’s alpha was calculated to determine the reliability of the scale [35]. The calculation of the relationship between the scores of the Perceived Benefit and Risk scales and Self-confidence was calculated with Spearman’s r correlation (rs). To determine the relationship between demographic and perceptual variables, non-parametric tests were used given the nature of the variables (ordinal) and the sizes of the sub-samples. Next, statistical tests such as the Mann-Whiney U (Z) were used to contrast the differences between the two groups and the Kruskal Wallis Test (similar to ANOVA), to contrast more than two groups.

To assess the reliability of the scales, an item-total analysis was performed. Tables 2 and 3 show the results of applying the maximum likelihood method with varimax rotation. The correlations indicate how each variable contributes to that dimension and is expressed through the Pearson correlation coefficient. High values of the coefficient (positive or negative) indicate that this certain variable contributes significantly to the creation of said dimension. When carried out on the Risk-Benefit scale, it was determined that items 3, 5, 12, 13 and 14 presented a low interrelation with the rest of the items on the scale, so they were removed from the questionnaire. The Cronbach’s alpha value of the scale after removing the five questions is 0.877. When performing the reliability analysis on the Self-Confidence Index scale before the FPDIP, all items showed a correlation with the total greater than 0.5, so no item was removed, and in this case, Cronbach’s alpha was 0.937 (See Tables 2 and 3).

The risk and benefit index in the FPDIP had a mean score of 3.13 out of 5, with a standard deviation (SD) of ±0.52, while, for self-confidence, it had a mean of 3.32 with a SD ±0.73. The range in RB ranges from a minimum score of 1.71 to a maximum of 4.39 and that of SC was between 1 and 1.48.

The resulting F2 dimension (Table 2) correlates strongly with variables that measure information about the patient’s relatives positively. Items 27, 28, 31 and 32 present r > 0.70. Item 4, with a lower correlation, is associated with this dimension since it directly involves family members (in this case, the person who answers the questionnaire). The F1 dimension includes the correlation with the variables that collect a direct interaction that has to do with the procedures. Also, most of the correlations are r > 0.70. And dimension F3 covers two variables that focus on the period after the intervention, as a conclusion of it concerning families. The correlations of items 42 and 43 with this dimension are r > 0.80.

Regarding Table 3, weaker correlations are observed between the dimensions and the variables that compose them. Dimension F2 is mainly made up of variables related to the exercise of the rights of family members to witness the intervention, as well as the benefits that it would bring to both family members and patients. However, the correlations here are between r|0.40 < r < 0.70|.There is a clearer correlation among the items of the F1 dimension: most of them refer to the positive consequences that the presence of relatives has on the performance of the professionals and their job satisfaction. And finally, dimension F3 is strongly correlated with the items that expose the right that both patients and their relatives must be present (r > 0.70).

Of the 120 professionals who responded to the survey, representing 22.18% of the population of medical and nursing personnel, 77.5% (n = 93) of them were women. 41.7% (n = 50) were between 36 and 46 years old, and only 19.5% (n = 24) were under 35 years of age. 63.3% (n = 76) of participants were nursing professionals and the work units with the most significant number of participants were the acute hospital ward with 33.3% (n = 40); socio-health centre, with 21.7% (n = 26) and the emergency department, with 12.5% n = 15). Regarding their training, 46.7% (n = 56) were graduates and more than half of the sample, 51.7% (n = 62) included a training speciality. It is relevant that the sample was composed mainly of professionals who worked in the acute care, social health and emergency units, so that it can be deduced that the answers obtained are the result of their own experiences, since in their professional areas IP is carried out frequently (See Table 4).

Only 45% (n = 54) of the sample invited a relative to witness an IP carried out on a family member, and of these, only 19.2% (n = 23) did it more than five times, there is a relationship between the fact that the professional was previously invited to FPDIP with the number of times the professional invited (X2 = 20.80; p < 0.001). In general, there is no association between years of experience and the number of times they allowed the FPDIP, highlighting that the majority (41%) of those who facilitated the FPDIP had 11 to 15 years of experience. On the other hand, there is a relationship between those who show greater self-confidence with expressing the desire for FPDIP in the Advance Care Planning.

There is a significant difference between the perceptions of the FPRB (F 3.67, p = 0.01) and FPSC (F 6.31, p < 0.001) indices and the opinion of who should decide on the FPDIP. The participants who presented a higher confidence index (Mean score: 3.6) considered that the person responsible for making the decision was the patient (44.2%; n = 53) and in contrast, those who presented lower Self-Confidence (Mean score: 2.9) indicated that the physician (34.2%) was responsible for the invitation. A similar result was obtained with the Risk-Benefit scale. Those participants who associated the invitation to a more significant benefit (Mean score: 3.26) considered that the decision had to be made by the patient, while those who identified a greater risk with the presence of family members (Mean score: 2.89) considered that the physician was the one supposed to invite to the FPDIP.

The survey questions from 44 to 50 were directly related to personal opinion and experience about the FPDIP. Most of the responses reported that they would not agree for a family member to be present while they were undergoing an IP, being more reluctant those who worked in emergencies. Additionally, it was found that regarding inviting the family, there is no significant difference between genders or work units, but there is a difference between the professions, physicians and nurses, according to the Mann-Whiney U (See Table 5).

The correlation between the Risk Benefit and Self-confidence variables is significant and with a moderate intensity of the relationship (Spearman’s rho of 0.58 and alpha < 0.001). In other words, there is a significant relationship although the strength of the association is moderate. In this case, as in the Twibell article, the perception of greater benefit is related to the perception of greater confidence of the professionals.

There is a significant relationship between the profession and the Risk Benefit (F = 6.10; mean = 1.49; p = 0.003) and self-confidence (F = 10.64; mean = 3.32; p < 0.001) and a greater perception of benefit (mean 3.48) than the medical group (mean 2.93), as well as greater self-confidence (mean 3.48) than physicians (mean 2.98).

There is no association between the FPRB and FPSC indices and the variables of gender, workplace, and professional experience. A trend is observed along with generational change, since younger professionals are more likely to allow PFDIP.

This study is a cross-sectional pilot test, but it has had a very low response rate (22.18%). The causes are unknown, it may be due to the long length of the questionnaire and its subsequent abandonment or discomfort and disagreement with the subject, for example.

There may be a gender bias, as only 27 men responded to the questionnaire, compared to 93 women.

This scientific article is the first step to explore the subject of FPDIP in adults, generating a tool that enables comparison with other evidence, from a health perspective and in Spanish language. More studies are needed on FPDIP.