Measuring red tape in a hospital setting: A survey experiment

Abstract

Public administration research is actively exploring alternatives for the General Red Tape (GRT) scale to measure red tape. Owing to increasing criticism on the GRT scale, scholars proposed the Three-Item Red Tape (TIRT) scale as an alternative. Using a repeated cross-sectional design, this article tests both scales in a before–after analysis of a major change in the organization of administration in a hospital. The results indicate that the GRT scale does not capture the resulting major change in red tape, which raises questions on the instrument's validity in a bottom-up research design within one organization. The TIRT scale, however, which measures red tape at the work environment level, does reflect the change in red tape but shows empirical weaknesses in its design. Additionally, by randomly assigning respondents to substantially different red tape definitions, this article shows that the red tape definition does not significantly impact respondents’ GRT ratings.

Points for practitioners

The predominantly used GRT scale is not able to capture an increase in red tape in a bottom-up intraorganizational research design in a hospital, which raises questions on the instrument's validity.

A more recent alternative for the GRT scale, more specifically the TIRT scale, captures the increase in red tape but shows empirical weaknesses.

The wording of the red tape definition does not impact respondents’ answers on the GRT scale.

Keywords

General Red Tape scale hospital public administration red tape survey experiment Three-Item Red Tape scale

Introduction

Recent empirical public administration research focuses mainly on the effects of red tape on organizational or individual factors and the drivers of red tape perceptions (see e.g. Jacobsen and Jakobsen (2018); Kaufmann and Tummers (2017); Migchelbrink and Van de Walle (2020); Oh and Lee (2020)). Scholars have demonstrated that reducing red tape leads to enhanced organizational performance, which emphasizes the importance of the concept in public and private policy-making (see e.g. De Jong and Van Witteloostuijn (2015); Walker and Brewer (2009)). To date, however, significant uncertainty remains in the literature on how to measure red tape. This article contributes to the existing red tape literature by examining two major methodological shortcomings using an experimental survey design.

First, most scholars employ the General Red Tape (GRT) scale, introduced by Rainey et al. (1995), to gain insight into red tape. Over the years, however, the GRT scale has received a substantial amount of criticism (see e.g. Bozeman and Feeney (2011); Luton (2007)), which resulted in scholars shaping new measures, including the Three-Item Red Tape (TIRT) scale of Borry (2016). One of the critiques is aimed at the definition used in the GRT scale. Scholars claim the definition carries a negative connotation, triggering respondents to give a more negative response on the scale (see e.g. Bozeman and Feeney (2011); Feeney (2012)). Moreover, Feeney (2012) finds ratings of perceived organizational red tape to be substantially influenced by the definition accompanying the GRT scale. Another frequently used red tape definition is the European Commission's (EC) conceptualization of red tape, which is considered to be more practice-based, and distinguishes rules from red tape more explicitly compared with the original GRT definition (European Commission, 2017). Using unique, primary survey data from a hospital and by randomly assigning one of the two definitions to respondents, this article tests whether the random presentation of the original GRT definition or the EC definition leads to variations in red tape ratings.

Second, to date, the predominant research approach of red tape scholars is to collect data from multiple organizations each represented by one manager to gain insight into organizational red tape at a given point in time, using a top-down oriented cross-sectional research design. Recently, however, leading red tape scholars identified this research approach as one of the caveats currently present in red tape research (see e.g. Kaufmann et al. (2020)). The predominantly used cross-sectional design does not allow researchers to assess the validity of red tape instruments. A repeated cross-sectional design, however, does provide the possibility to examine the validity of the instrument, especially when a fundamental transformation of the administrative processes changes red tape levels, by testing whether this change is reflected in the instrument's results. In this study, the major transformation in administrative processes was the implementation of an integrated Electronic Health Records (EHR) system, which replaced various fragmented software packages on a single day, and therefore provides a clear cut-off in time for the analyses. Hence, the second aim of this article is to examine two red tape instruments (i.e. GRT and TIRT) over time and their sensitivity to major changes in red tape using a bottom-up oriented, repeated cross-sectional research design in a single organization.

Theory

In his theoretical contribution to red tape research, Bozeman (1993) conceptualized red tape as “rules, regulations, and procedures that remain in force and entail a compliance burden for the organization but have no efficacy for a rule's functional object”. To examine the differences in perceived red tape between private and public managers, Rainey et al. (1995) used this definition to shape the General Red Tape (GRT) scale. The GRT measure is a scale-based answer to the question, “If red tape is defined as burdensome rules and procedures that have negative effects on the organization's effectiveness, how would you assess the level of red tape in your organization?” Respondents are asked to indicate the level of perceived organizational red tape on a 0 to 10 scale in which 0 corresponds to almost no red tape and 10 represents a great deal of red tape, with the purpose of gaining insight into perceived levels of organizational red tape.

Since its introduction, mostly owing to its ease of application and generic nature, the GRT scale has been used in a wide variety of domains. Bozeman and Feeney (2011) argue the GRT scale is one of the best examples of consistent, cross-research application of an empirical measure in public administration research. The scale has been applied in surveys administered to public, private and non-profit managers, health and human services departments, city managers, and so forth. Yet, empirical red tape research in healthcare organizations is, to date, fairly limited (Henderson and Borry, 2020; Van Loon, 2017), especially in hospital settings.

Over the years, scholars started criticizing the GRT scale. Luton (2007) argues that respondents first have to decide which rules are burdensome and subsequently if those burdensome rules affect organizational effectiveness in a negative way. Respondents thus have to make two judgments before making the requested evaluation on a single indicator scale. From the work of Bozeman (Bozeman, 2000; Bozeman and Feeney, 2011), Borry (2016) argues that only rules that are burdensome, unnecessary, and ineffective are considered red tape and that the single indicator GRT scale may not be able to capture this complexity of the red tape concept. In a critical article on red tape research, Bozeman (2012) stated that scholars have chosen simplicity at the cost of realism.

Borry (2016) therefore argues there is a need for a multiple indicator scale to counter previously mentioned critiques. She proposes the TIRT scale, which is composed of three rule characteristics distinguishing rules from red tape: burden, effectiveness, and necessity. The TIRT scale is a multiple indicator scale asking respondents to answer the question, “How would you describe policies and procedures in your work division between the following opposite characteristics?” These characteristics range from not burdensome to burdensome, necessary to unnecessary, and effective to ineffective.

The TIRT scale has several advantages over the GRT scale. First, because of the multiple indicator scale scholars gain more insight into the extent that rules are considered to be red tape. This distinction can be of fundamental importance in healthcare. Although rules and procedures may be burdensome, they can be necessary in safeguarding patient safety and quality of care (Van Loon, 2017). Second, the TIRT scale asks respondents to evaluate red tape in their specific work environment in contrast to the entire organization, which is in line with the recent evolution towards multiple stakeholder measurement. Although this could lead to significant differences between GRT and TIRT outcomes, recent articles have shown TIRT levels to be very close to GRT levels but also discovered a weak internal consistency of the TIRT scale as a construct variable (Kaufmann et al., 2020, 2021). The conceptual similarity between the GRT and TIRT scales could be explained by the fact that the TIRT scale was built on the same theoretical foundations as the GRT scale. It is because of this similarity we preferred the TIRT scale over other recently introduced red tape measures (such as the Job-Centered Measure of Van Loon et al. (2016)).

The aforementioned critiques of leading red tape scholars lead to uncertainty among researchers whether the GRT scale is a valid instrument to measure red tape, and whether the TIRT scale is a viable alternative (Kaufmann et al., 2019). Since the predominant approach in red tape research is to collect red tape ratings of multiple organizations, each represented by a single manager at a given point in time, it is difficult to assess the validity of the GRT scale. When multiple measurements — preferably among multiple stakeholders within each organization — would be performed in the same organization(s) over time, one could assess the validity of the GRT measurement by manipulating the amount of red tape experienced by respondents and analyse whether this change is reflected in GRT scale's ratings (Pandey and Marlowe, 2014). The aim of this article is to simultaneously investigate the validity of the GRT and TIRT scales using a before–after analysis of a major change in the administrative processes of a hospital in a repeated cross-sectional design. In contrast to Pandey and Marlowe (2014), who changed red tape levels artificially using a vignette-approach in an experimental setting, this article examines a real-life change in red tape owing to the transformation of the administrative processes of a hospital avoiding potential hypotheticality bias. This major change was the replacement of various software packages with an integrated EHR system. The fragmented software packages mainly consisted of free text entry spaces while the new EHR used check boxes in order to promote structured data gathering. These check boxes focused on a wide range of potential risk factors resulting in more extensive reporting compared with the free text entry in which only the most relevant findings were reported. Furthermore, the new EHR consisted of numerous features the previous software packages did not (e.g. calculation of early warning scores), resulting in more extensive reporting. Therefore, based on the abovementioned information, workshops with hospital staff, and observations at various departments in the hospital both before and after the implementation of the EHR, we argue red tape levels will increase in the hospital and, consequently, red tape ratings on the GRT and TIRT scales are expected to increase as well. This leads to the following hypothesis:

H₁: A major increase in red tape leads to an increase in GRT and TIRT ratings.

Opposite to the vast body of red tape research, mainly using the original GRT scale of Rainey et al. (1995), the EC employs a different approach to measure red tape in the Standard Cost Model. The model is based on two types of administrative costs: the business-as-usual costs and the administrative burdens.¹ Business-as-usual costs are the costs resulting from and time spent on information obligations that would also be done in the absence of procedures and legislation. In other words, these costs are the costs of necessary and effective rules and should therefore not be considered as red tape. The administrative burdens, on the other hand, are the costs resulting from activities that are done solely because of a legal obligation, and are therefore deemed unnecessary, ineffective, and burdensome (European Commission, 2017). The EC thus uses a substantially different definition to define red tape in contrast to the majority of red tape scholars: “Administrative burdens represent the time and costs incurred in meeting legal obligations to provide information on your activities, either to public or private authorities, which are done solely because of this legal obligation” (European Commission, 2017). The EC conceptualization of red tape is widely used and accepted among policy makers and has been taken over by EU member states, the OECD, and scholars (e.g. Arendsen et al. (2014); OECD (2007); Poel et al. (2014)).

In a measurement experiment using three alternative definitions, Feeney (2012) concludes definition wording is of significant importance when using the GRT scale as it affects respondents’ evaluations. This experiment distinguishes itself from Feeney's research in three ways. First, this study is performed in a hospital context rather than in government agencies. While local government managers’ daily business is administration, hospital staff are mainly focused on the provision of health care. Second, this study uses a bottom-up in contrast to a top-down approach by including multiple hospital staff members with different job contents within one organization. Third, while Feeney used three theoretical red tape definitions as alternatives for the original GRT definition, this study employs a practice-based alternative to define red tape. By randomly assigning respondents to either the GRT or the EC definition, this article examines whether a perceptual difference can be observed between the two alternatives with the final hypothesis:

H₂: Variations in perceived levels of organizational red tape stem from a difference between the GRT and EC definition.

Methods

Study and survey design

This article employs primary survey data collected in a Belgian general hospital with five campuses. The hospital accommodates approximately 3000 health professionals and accounts for more than 650,000 patient consultations each year. This study uses a repeated cross-sectional design to test the aforementioned hypotheses. The reason for opting for multiple measurements in time is to be able to accurately assess the sensitivity of the GRT and TIRT scales to changes in red tape. The hospital participating in this study was to undergo a major change in its administrative processes. This major change was the implementation of an integrated EHR, which had to replace multiple software packages. The hospital used a rather unique implementation strategy in which the transition from the fragmented software to the integrated EHR took place in a single day, whereas most hospitals opt for gradual implementations of the different modules over time.

A red tape survey was administered at three points in time to conduct a before–after analysis. The first measurement was conducted one month before the implementation of the EHR to avoid other major organizational changes to bias the results (e.g. different work protocols, staffing, etc.). Following the healthcare technology impact and acceptance literature, the second measurement took place 11 months after implementation for two reasons. First, by waiting almost a year we were able to avoid measuring the irritation inherent to a substantial organizational change that could bias red tape perceptions. Second, after using the system for one year, learning effects and the inherent dip in performance should be minimal as hospital staff have gained sufficient experience with the new system. The final measurement was conducted 14 months after implementation. The relatively short time span between the second and third survey was chosen on purpose since the final survey served as a robustness check for the results obtained in the second survey. Since no changes in organizational processes were carried out during this period, the results of the second and third survey should thus be highly comparable.

The survey was administered to all types of health professionals and generally consisted of three sections: professional profile, red tape perceptions, and respondent characteristics. The professional profile section asked respondents to indicate their profession, experience on the job, amount of hours worked in a week, and campus. The red tape section focussed on the amount of time spent on, respectively, administration, communication with colleagues, and patient interaction. Furthermore, this section included the GRT and TIRT scales. A detailed overview of the red tape sections in the three surveys is shown in Figures 1 –3. In the first and second survey, each respondent was randomly assigned to either the original GRT scale or the EC GRT scale by the survey software package. In order to compare the TIRT and GRT scales, this article uses a 0–10 scale for the TIRT scale instead of the 5-point scale proposed by Borry (2016). Note that in the surveys containing both the GRT and TIRT scales, the TIRT scale, which was identical to all respondents, precedes the GRT scale with alternating definitions to avoid biased TIRT ratings by the definition respondents were presented with when completing the GRT scale. Because of this reason, and to achieve sufficient response and power for the analyses, the order in which the scales appeared was not randomized. The final section of the surveys contained questions concerning respondent characteristics, such as age, gender, and well-being at work.

Survey development and distribution

In order to translate the TIRT scale and both versions of the GRT scale, this article used the combined translation technique proposed by Cha et al. (2007). To collect the data, electronic surveys were sent out to the various stakeholder groups (i.e. physicians, head nurses, nurses, paramedical staff, and secretaries). Representatives of each stakeholder group were involved in the development of the survey instrument during workshop sessions organized for each group separately. Before distributing the survey among physicians and hospital staff, the final survey instrument was evaluated and approved by two independent ethical committees. The first survey was completed by 383 respondents (13% of hospital staff), the second by 836 respondents (29%), and the third by 157 respondents (5%). The low response rate of the final measurement can be explained by the limited time interval between the second and the final survey. Owing to these relatively low response rates, a sample representability check was performed by comparing the composition of the samples to the composition of the population. As shown in Appendix 1 (see online supplemental material), the samples are a reliable representation of the hospital population. To interpret survey results, multiple workshops and interviews with stakeholder representatives were performed in the weeks after survey completion. The entire study was conducted between 2018 and 2019.

TIRT scale

As the TIRT scale is a relatively new instrument in public administration research, and especially in a healthcare context, this article first evaluated the TIRT scale using confirmatory factor analyses (CFA) to assess construct validity. Cronbach's alpha was employed to check internal consistency of the TIRT construct.

GRT and TIRT validity test

The primary aim of this article is to investigate whether the GRT and TIRT scales capture a change in red tape levels owing to a major change in administrative processes. Therefore, we will test whether red tape levels did in fact increase as a consequence of this change with a set of self-reported control variables in the survey, such as the proportion of time spent on administration relative to communication with colleagues and patient interaction, well-being at work variables, amount of hours worked on a weekly basis, and comparative red tape perception questions on subtasks. The well-being at work variables include job satisfaction, motivation, job qualification, workload, and risk of burnout. These variables were included since multiple studies have shown the significant relationship between well-being at work variables and red tape perceptions (Feeney, 2012; Feeney and Bozeman, 2009; Giauque et al., 2012; Steijn and van der Voet, 2019). All well-being at work variables were measured using a 7-point Likert scale ranging from very low to very high.

To check on a lower level whether red tape levels changed, we employed a specific set of core administrative work tasks and asked respondents for each task to indicate whether they perceived more or less red tape compared with the situation before the new EHR. The identified administrative tasks include reporting, registrations, orders (for labs, scans, etc.), administration concerning medication (e.g. ordering, admission registration), and anamnesis. Furthermore, we added tasks on which the new EHR should not have an impact as a control mechanism to check respondents’ choice patterns (i.e. briefings and quality control). The comparative red tape questions concluded by asking respondents whether they experienced, in general, more or less red tape in their job activities compared with the situation before the new EHR. Finally, extensive workshops with hospital staff and an extensive series of observations were performed at various, carefully selected departments in the hospital both before and after the implementation of the EHR.

Second, we will examine whether the GRT scale and/or the TIRT scale pick up the presumed change in red tape levels. To statistically test whether GRT and TIRT results significantly change over time, we used t-tests.

GRT versus EC conceptualization of red tape

As previously mentioned, we hypothesize the perceptual difference between the GRT and EC definitions influences red tape evaluations. The operationalization of both versions is shown in Appendix 2 (see online supplemental material). To test whether there is a definition effect, leading to variations in ratings on the 0–10 scale, each respondent was randomly assigned to one of the two definitions. First, a t-test was performed to test the equality of means of GRT and EC levels. Second, regression models were estimated using the ordinary least squares (OLS) estimation technique while controlling for profession, age, gender, campus, experience, and the timing of the survey (i.e. before or after the EHR implementation). To make sure no structural difference between the GRT and EC respondents exists, leading to different perceptions on red tape levels not attributable to the displayed definition, we employed the percentage of time spent on administration (identical to all respondents) as a check for balance between the two groups.

Results

The descriptive results of the three surveys, presented in Table 1, show that on average respondents perceive high levels of red tape in the hospital when using the GRT scale. With averages above 7, the GRT levels in this setting are among the highest observed in red tape research (Borry, 2016; Kaufmann and Feeney, 2012; Welch and Pandey, 2006). This stresses the importance of the red tape problem in hospitals. Remarkably, 30–35% of work-related time was reported to be spent on administration, while only 41–46% of time is spent on patient care and communicating to patients’ families.

Table 1.

Descriptive statistics.

		$T_{0}$					$T_{1}$					$T_{2}$
		N	Mean	Std Dev.	Min	Max	N	Mean	Std Dev.	Min	Max	N	Mean	Std Dev.	Min	Max
GRT		383	7.2167	1.7966	0	10	836	7.1734	1.7855	0	10	157	7.0828	1.8535	0	10
	GRT definition	191	7.1257	1.8764	0	10	418	7.1938	1.8068	2	10	157	7.0828	1.8535	0	10
	EC definition	192	7.3073	1.7136	1	10	418	7.1531	1.7657	0	10	/	/	/	/	/
PercPatients		316	46.6118	19.8981	0	94.2857	796	41.3960	21.4414	0	100	133	42.8217	18.1029	0	87.5000
PercComm		316	23.7750	12.8905	1.3333	73.5294	796	24.0480	13.7983	0	83.3333	133	22.6318	12.6219	0	64.5161
PercAdmin		316	29.6132	15.1288	2.8571	86.9565	796	34.5561	17.2788	0	100	133	34.5465	16.8250	0	93.5484
TIRT		383	3.6049	1.8680	0	10	/	/	/	/	/	157	4.1274	1.9834	0	10
	Burdensome	383	4.8695	2.6186	0	10	/	/	/	/	/	157	5.2357	2.6120	0	10
	Unnecessary	383	2.6162	2.2305	0	10	/	/	/	/	/	157	3.1147	2.2100	0	10
	Ineffective	383	3.3290	2.3516	0	10	/	/	/	/	/	157	4.0318	2.4532	0	10

TIRT scale

To check the internal consistency of the TIRT scale as a construct variable, we first calculated Cronbach's alpha (0.67 at T₀ and 0.75 at T₂). With a value at T₀ below and at T₂ slightly above the threshold value of 0.70, the internal consistency of the TIRT construct is relatively poor. The reason for this finding can be found in the correlation coefficients of the three indicators. The burden indicator is weakly correlated with the unnecessary (0.19 at T₀ and 0.38 at T₂) and ineffective (0.30 at T₀ and 0.47 at T₂) indicators, while the unnecessary and ineffective indicators are correlated more strongly (0.73 at T₀ and 0.67 at T₂). Since internal consistency of the TIRT scale in the first survey is too low and model convergence was not achieved, the CFA, presented in Figure 4, only employs the data of T₂.

The results at T₀ suggest that the three characteristics of red tape are not suited to being treated as one construct variable in the context of this research. This is in line with recent findings of Kaufmann et al. (2020) and Hattke et al. (2020). While the results at T₂ show an acceptable Cronbach's alpha, the variance explained in the burden indicator is low (21%). When constructing two variables instead of one construct variable, Cronbach's alpha of the unnecessary–ineffective construct increases to 0.84 at T₀ and 0.80 at T₂. As already mentioned, it is likely in this context that certain administrative tasks (e.g. allergies registration) are considered as burdensome, while they are necessary in safeguarding patient safety and quality of care. Consequently, these results seem to indicate that at least two separate constructs are required to measure red tape in a valid way.

GRT and TIRT analyses over time

To examine the second hypothesis in this article, we will examine the evolution in the GRT and TIRT scales, after which we will demonstrate red tape levels did increase in the organization as a consequence of the new EHR using multiple control variables. Table 2 shows the impact of the implementation on the red tape variables in the first panel, and the impact on the control variables in the second panel. The results indicate that the GRT averages do not vary as much over time as expected. The differences between the averages before and after the implementation are therefore not significant (p < 0.1). Since we employed two different red tape definitions accompanying the scale (i.e. GRT and EC definition), the analysis was repeated for each version separately. However, this leads to the same conclusion.

Table 2.

Before–after analysis.

	T₀	T₁	p valueT₀ vs T₁	T₂	p valueT₀ vs T₂
Red tape variables
GRT scale	7.22	7.17	.6590	7.08	.4143
GRT definition	7.13	7.19	.6901	7.08	.8133
EC definition	7.31	7.15	.2885	/	/
TIRT scale	3.60	/	/	4.13	.0039
Burdensome	4.87	/	/	5.24	.1403
Unnecessary	2.62	/	/	3.11	.0184
Ineffective	3.33	/	/	4.03	.0019
Control variables
% of time spent on administration	29.61	34.56	.0000	34.55	.0024
Job satisfaction	5.07	4.59	.0000	4.63	.0001
Motivation	5.60	5.26	.0000	5.14	.0000
Job qualification	5.53	5.36	.0049	5.27	.0032
Risk of burnout	3.76	4.02	.0061	4.07	.0315
Workload	5.43	5.47	.5762	5.36	.4665
Hours worked(weekly)	37.81	38.23	.5103	36.02	.0451

When looking at the TIRT scale a different conclusion can be made. As shown in Table 2, the red tape average measured by the TIRT scale does increase significantly. When repeating the same analysis for each of the TIRT indicators, it can be observed that both the ineffective and unnecessary indicators increase significantly but the increase in the burdensome indicator is not statistically significant. From the workshops after survey completion, it could be derived that hospital staff had to complete more administrative tasks than before the implementation. However, using the fragmented software, hospital staff endured a higher mental burden owing to the free text entry as they had to think of all relevant observations themselves. The EHR, on the other hand, was less mentally demanding since it required them to check all possible observations one by one. While the administrative tasks became more burdensome in time, they got less mentally burdensome to complete, which partly offsets the time increase in the burdensome indicator. In other words, while procedures and rules are perceived as more unnecessary and definitely more ineffective – mainly owing to the fact that several check boxes are not applicable to the patient and the addition of new features to the EHR not deemed necessary to safeguard quality of care – the overall burden of these tasks did not increase significantly.

The contrasting results of both scales raises questions on which scale gives the most reliable representation of the impact of the EHR on red tape in the hospital. Therefore, several control variables were added in the analyses to be able to conclude on this dilemma. As shown in panel 2 of Table 2, the percentage of work-related time spent on administrative tasks increases significantly from 30% to 35% in both follow-up measurements. Furthermore, each of the well-being at work variables deteriorates significantly over time. An exception in the well-being at work variables is workload, which remains at the same level as well as the hours worked in a week, which even decrease slightly in the final measurement. This would explain why the burdensome indicator of the TIRT scale did not increase significantly over time since hospital staff do not experience a higher overall burden in their job, but have the feeling they have to perform less necessary and effective tasks.

To supplement the aforementioned control variables, comparative red tape questions were used to be able to judge whether red tape levels did in fact increase or not. As shown in Table 3, respondents indicated experiencing significantly more red tape in the key administrative work tasks. To check whether respondents did not indicate increased red tape in all tasks out of frustration with the new EHR, we added two tasks that should not be impacted by the new EHR (i.e. briefings and quality control). As presented in Table 3, the averages of both tasks are around 4, indicating red tape levels remained the same and the above answers can be deemed valid. Finally, an overall comparative red tape question, which can be interpreted as a proxy for the evolution in red tape, also indicates red tape did in fact increase in the organization with almost 75% of respondents experiencing this increase. The observations, interviews, and workshops with hospital staff confirmed these results.

Table 3.

Comparative red tape questions.

T₂ compared to T₀
Administrative tasks
Reporting	5.16
Registrations	5.52
Orders (e.g. labs, scans, etc.)	5.01
Medication registration	5.13
Anamnesis	5.21
Tasks on which the EHR should not have an impact
Briefings	4.06
Quality control	4.38
Overall comparative red tape question
Overall, do you perceive more or less red tape compared to the situation before the implementation of the EHR?	5.49

Note: 7-point Likert scale from ‘a lot less’ to ‘a lot more’.

As demonstrated by the results of the control variables and the comparative red tape questions, it is clear that the implementation of the new EHR did not lead to more formalization, but instead led to an increase in organizational red tape. This sudden, substantial increase in red tape levels offers the opportunity to test the validity of both the GRT and the TIRT scales by examining whether this increase is reflected in the results of each scale. From the analyses of the GRT and TIRT scales, it is clear that only the TIRT scale increases significantly over time and should thus be considered as the more valid instrument of the two. The most plausible explanation for this finding is that the GRT scale does not measure red tape accurately as it is too difficult for staff members to evaluate it in the entire organization. In this case, the GRT scale does not detect the change in organizational red tape owing to the bottom-up approach compared with the frequently used top-down approach. Managers could be in a better position to judge red tape in the entire organization in contrast to staff members. This would also explain the large differences in GRT and TIRT averages within each measurement, which have not been observed in the past, as the TIRT scale focuses on work environment red tape and the GRT scale on organizational red tape (Borry, 2016; Kaufmann et al., 2020, 2021). The GRT scale could thus be an instrument that is ultimately suited to conducting interorganizational research using a top-down approach, but not appropriate for intraorganizational research following the bottom-up approach.

GRT versus EC conceptualization

To examine whether respondents perceive different levels of red tape depending on the definition used in the GRT scale, each respondent was assigned randomly to one of two definitions. The unbiased percentages of time spent on administration, which preceded the randomization in the survey, were used to perform a check for balance with t-tests between the two groups. As Appendix 3 (see online supplemental material) demonstrates, the group that was allocated to the EC definition is highly comparable with the group allocated to the GRT definition.

To investigate whether a different definition leads to different results on the single indicator scale, a t-test was performed in each of the two measurements. The results of the t-tests clearly indicate no significant difference in average ratings between the GRT and EC definitions (p < 0.1). This analysis was repeated for each stakeholder group separately in Appendix 4 (see online supplemental material), which leads to an identical conclusion. Additionally, we examined the data further by estimating two regression models using the ordinary least squares (OLS) estimation technique to control for professional and demographic characteristics, and the timing of the survey. The first model controls for profession, gender, age, campus, and the timing of the survey (before or after EHR implementation). The second model additionally controls for experience. The estimated models, shown in Table 4, clearly indicate there is no significant definition effect, even at the significance level of 0.1. This would suggest that although the two definitions differ significantly in wording and emphasis, respondents do not alter their red tape perceptions based on the definition provided to them. Post-survey workshops indicated respondents already have an overall idea of what red tape is and to what extent it affects their clinical practice, and that because of this general belief the definition provided to them did not influence red tape perceptions. The results of the regression analyses also provide additional support for the previously discussed finding that there is no significant difference over time in GRT averages when controlling for respondents’ professional and demographic characteristics and the definition provided to them.

Table 4.

Regression models using OLS.

Variables	(1) GRT	(2) GRT
Profession
Nurse	.0159	−.0447
	(.1558)	(.1594)
Paramedical staff	−.6767***	−.7453***
	(.2277)	(.2289)
Physician	−.3919*	−.3701*
	(.2083)	(.2097)
Secretary	−.3797	−.4670*
	(.2558)	(.2581)
Others	−.4818*	−.4683*
	(.2627)	(.2625)
Gender
female	−.1651	−0.1678
	(.1427)	(.1435)
Age
26–35	.2154	.1389*
	(.1649)	(.1836)
36–45	.0168	−.1893
	(.1721)	(.2019)
46–55	.01407	−0.2870
	(.1743)	(.2185)
Above 55	−.0388	−0.4707*
	(.2136)	(.2730)
Campus
B	−.1809	−.1568
	(.1404)	(.1408)
C	.2363*	.2101
	(.1387)	(.1380)
D	.2214	.2558
	(.1695)	(.1706)
E	.3176***	.3110***
	(.1132)	(.1120)
Experience
1–5 years		.0285
		(.2288)
6–10 years		.1835
		(.2449)
11–20 years		.4179*
		(.2434)
21–30 years		.4039
		(.2664)
Above 30 years		.7016
		(.3080)
Timing
After implementation(T₁)	−.2910	−.0646
	(.1132)	(.1144)
Definition EC	.0193	.0171
	(.1017)	(.1015)
Constant	7.2454***	7.2303***
	(.2655)	(.3044)
Observations	1219	1219
R-squared	.0325	.0414

Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1.

While not being the scope of this study, the results of the regression analyses also indicate significant differences in the GRT ratings between the different stakeholders in the hospital. Paramedical staff, physicians, and ‘others’ experience significantly less red tape compared with head nurses. When controlled for experience as well, this is also the case for secretaries. Furthermore, the results indicate that perceived red tape decreases with age, but increases with experience. This can be explained by the relatively high retraining rates towards the healthcare sector in the Belgian labor market. In our samples at T₀ and T₁, for example, 23% and 31% of the respondents older than 55 have less than 10 years of experience.

Conclusion

This article examines both the GRT shortcomings and the potential of the TIRT scale as an alternative using a survey experiment in a Belgian general hospital. First, we tested the validity of both scales using a before–after analysis in a repeated cross-sectional research design by examining whether each scale is able to capture a major increase in red tape. Analyses showed that the change in red tape is only reflected in the TIRT scale, while the average of the GRT scale remained constant. The GRT scale, in this context, thus fails to act as an instrument to measure the impact of organizational change or policy measures on organizational red tape. A possible reason for this finding could be the intraorganizational bottom-up approach of this study in contrast to the more frequently used interorganizational top-down approach.

Second, we examined whether the definition used to describe red tape leads to variations in perceived levels of red tape. By randomly assigning either the original GRT definition or the EC's conceptualization of red tape to each respondent, we are able to conclude that although the two definitions differ in wording and emphasis, respondents reflect on the same underlying red tape concept when evaluating red tape in their organization.

Overall, we can conclude the GRT scale is not sensitive to changes in organizational red tape and the definition accompanying the scale does not influence results. The results seem to advocate for a multiple indicator scale, preferably one in which the term ‘red tape’ is not mentioned since respondents tend to make a cognitive shortcut to their own belief of what red tape means. Further longitudinal and experimental research is needed to examine the validity of the GRT scale and its alternatives in a multitude of research contexts. Furthermore, we discovered the internal consistency of the TIRT construct is relatively poor. Therefore, in future red tape research in a healthcare context, the Job-Centered Measure of Van Loon et al. (2016) should be explored to examine red tape perceptions and as an instrument to measure the impact of organizational change on red tape.

Supplemental Material

sj-docx-1-ras-10.1177_00208523211073498 - Supplemental material for Measuring red tape in a hospital setting: A survey experiment

Supplemental material, sj-docx-1-ras-10.1177_00208523211073498 for Measuring red tape in a hospital setting: A survey experiment by Janis Luyten and Wim Marneffe in International Review of Administrative Sciences

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Janis Luyten

Wim Marneffe

Supplemental material

Supplemental material for this article is available online.

Notes

Janis Luyten, PhD in Business Economics, is postdoctoral researcher at the Faculty of Business Economics, Hasselt University, Belgium. His research fields are public administration, health economics, and public economics.

Wim Marneffe is Professor of Business Economics at the Faculty of Business Economics, Hasselt University, Belgium. His research fields are public administration, law and economics, health economics, and public economics.

References

Arendsen

Peters

ter Hedde

, et al. (2014) Does e-government reduce the administrative burden of businesses? An assessment of business-to-government systems usage in the Netherlands. Government Information Quarterly 31(1): 160–169.

Borry

(2016) A new measure of red tape: Introducing the three-item red tape (TIRT) scale. International Public Management Journal 19(4): 573–593.

Bozeman

(1993) A theory of government ‘red tape’. Journal of Public Administration Research and Theory: J-PART 3(3): 273–303.

Bozeman

(2000) Bureaucracy and Red Tape. Englewood Cliffs, NJ: Prentice Hall.

Bozeman

(2012) Multidimensional red tape: A theory coda. International Public Management Journal 15(3): 245–265.

Bozeman

Feeney

(2011) Rules and Red Tape: A Prism for Public Administration Theory and Research. Armonk: Sharpe.

Cha

Kim

Erlen

(2007) Translation of scales in cross-cultural research: Issues and techniques. Journal of Advanced Nursing 58(4): 386–395.

De Jong

Van Witteloostuijn

(2015) Regulatory red tape and private firm performance. Public Administration 93(1): 34–51.

European Commission (2017) Better Regulation Toolbox. Brussels: European Commission.

10.

Feeney

(2012) Organizational red tape: A measurement experiment. Journal of Public Administration Research and Theory 22(3): 427.

11.

Feeney

Bozeman

(2009) Stakeholder red tape: Comparing perceptions of public managers and their private consultants. Public Administration Review 69(4): 710–726.

12.

Giauque

Ritz

Varone

, et al. (2012) Resigned but satisfied: The negative effect of public service motivation and red tape on work satisfaction. Public Administration 90(1): 175–193.

13.

Hattke

Hensel

Kalucza

(2020) Emotional responses to bureaucratic red tape. Public Administration Review 80(1): 53–63.

14.

Henderson

Borry

(2020) The emotional burdens of public service: Rules, trust, and emotional labour in emergency medical services. Public Money and Management. doi: 10.1080/09540962.2020.1831180

15.

Jacobsen

Jakobsen

(2018) Perceived organizational red tape and organizational performance in public services. Public Administration Review 78(1): 24–36.

16.

Kaufmann

Feeney

(2012) Objective formalization, perceived formalization and perceived red tape. Public Management Review 14(8): 1195–1214.

17.

Kaufmann

Tummers

(2017) The negative effect of red tape on procedural satisfaction. Public Management Review 19(9): 1311–1327.

18.

Kaufmann

Ingrams

Jacobs

(2020) Rationale and process transparency do not reduce perceived red tape: Evidence from a survey experiment. International Review of Administrative Sciences. doi: 10.1177/0020852320966037

19.

Kaufmann

Ingrams

Jacobs

(2021) Being consistent matters: Experimental evidence on the effect of rule consistency on citizen red tape. The American Review of Public Administration 51(1): 28–39.

20.

Kaufmann

Taggart

Bozeman

(2019) Administrative delay, red tape, and organizational performance. Public Performance and Management Review 42(3): 529–553.

21.

Luton

(2007) Deconstructing public administration empiricism. Administration and Society 39(4): 527–544.

22.

Migchelbrink

Van de Walle

(2020) Increasing the cost of participation: Red tape and public officials’ attitudes toward public participation. International Review of Administrative Sciences. doi: 10.1177/0020852320942311

23.

OECD (2007) Cutting Red Tape: Comparing Administrative Burdens Across Countries. Paris: OECD Publications.

24.

Lee

(2020) External control mechanisms and red tape: Testing the roles of external audit and evaluation on red tape in quasi-governmental organizations. International Review of Administrative Sciences. doi: 10.1177/0020852320974097

25.

Pandey

Marlowe

(2014) Assessing survey-based measurement of personnel red tape with anchoring vignettes. Review of Public Personnel Administration 35(3): 215–237.

26.

Poel

Marneffe

Bielen

, et al. (2014) Administrative simplification and economic growth: A cross country empirical study. Journal of Business Administration Research 3(1): 45.

27.

Rainey

Pandey

Bozeman

(1995) Research note: Public and private managers’ perceptions of red tape. Public Administration Review 55(6): 567–574.

28.

Steijn

van der Voet

(2019) Relational job characteristics and job satisfaction of public sector employees: When prosocial motivation and red tape collide. Public Administration 97(1): 64–80.

29.

Van Loon

(2017) From red tape to which performance results? Exploring the relationship between red tape and various dimensions of performance in healthcare work units: Red tape and performance. Public Administration 95(1): 60–77.

30.

Van Loon

Leisink

Knies

, et al. (2016) Red tape: Developing and validating a new job-centered measure. Public Administration Review 76(4): 662–673.

31.

Walker

Brewer

(2009) Can management strategy minimize the impact of red tape on organizational performance? Administration and Society 41(4): 423–448.

32.

Welch

Pandey

(2006) E-government and bureaucracy: Toward a better understanding of intranet implementation and Its effect on red tape. Journal of Public Administration Research and Theory 17(3): 379–404.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB