Attitudes of dental professional staff and auxiliaries in Riyadh, Saudi Arabia, toward disclosure of medical errors.

AIM: To collect empirical data on the attitudes of dental professionals and dental auxiliaries in Riyadh, Saudi Arabia, regarding the disclosure of medical errors.
METHODS: A cross-sectional study was conducted, involving the administration of a questionnaire to a sample of 586 participants recruited from over 10 government and private dental institutions in Riyadh between August 2015 and January 2016. The questionnaire collected information regarding participant opinions on (a) personal beliefs, norms, and practices regarding medical errors, (b) the nature of errors that should be disclosed, and (c) who should disclose errors.
RESULTS: Most (94.4%) participants preferred that medical errors should be disclosed. However, personal preferences, perceptions of the norm and current practices with respect to which type (seriousness) of error should be disclosed were inconsistent. Only 17.9% of participants perceived that it was the current practice to disclose errors resulting in "Major harm". Over 68% of respondents reported a personal belief, a perception of the norm and a perception of current practice that errors should be disclosed by the erring dentist. Participants at government institutions were more likely to disclose errors than those at private institutions. There were also significant differences in the responses with respect to gender, age, and nationality. The implications for the development of guidelines to help Saudi dentists adopt ethical courses of action for the disclosure of errors are considered.
CONCLUSIONS: (1) The majority of participants personally believed that errors should be disclosed, (2) there was little agreement between participant personal beliefs and perceptions of the norm and practice with respect to which type of errors should be disclosed, (3) there was strong agreement that the erring dentist is responsible for reporting errors, and (4) the attitudes of the participants varied with respect to type of institution, age, gender, and nationality.

Introduction

The most cited definition of a medical error is “An act of omission or commission in planning or execution that contributes or could contribute to an unintended result.” This definition includes “the key domains of error causation (omission and commission, planning and execution), and captures faulty processes that can and do lead to errors, whether adverse outcomes occur or not” (Grober and Bohnen, 2005). Historically, medical errors were rarely disclosed. However, more recently, with the implementation of professional codes of ethics, disclosure of medical errors in the healthcare setting has been reinforced to prevent or reduce harm to patients and their families (Chafe et al., 2009, Ozar and Sokol, 2002, Ghazal et al., 2014, Williams, 2012). Non-disclosure of a medical error is now considered a violation of ethical principles and can lead to litigation (Rosner et al., 2000).

The disclosure of medical errors improves the quality of the healthcare system and helps to prevent future errors (Ghazal et al., 2014). Patient response to and consequences of medical errors greatly influence the attitudes of healthcare providers. It is generally accepted that full disclosure of a medical error is necessary only if there has been an adverse event which has caused harm to a patient (Ghazal et al., 2014). In situations where no harm or adverse event has occurred, disclosure may not be obligatory (Elder et al., 2006) as it may unnecessarily increase patient stress and anxiety. Gallagher et al. (2009) reported that some physicians believe that if patients do not enquire then error disclosure is unnecessary.

Many factors may influence the decision of healthcare providers to disclose medical errors. According to the conceptual model conceived by Fein et al. (2005), the most important influences on the decision to disclose a medical error fall into four categories: (a) provider factors, including perceived professional responsibility, (b) patient factors, including a desire for information, (c) error factors, including the level of harm to the patient, and (d) institutional culture, including the perceived tolerance for error by healthcare professionals.

Birks, 2014, Ghazal et al., 2014 proposed a set of guidelines for the disclosure of medical errors, citing conceptual reasons, such as the duty of candor, respect for autonomy, the imperative principle of truth-telling, the principles of beneficence and non-maleficence, and the deontology or Kantian obligation based theory. Healthcare providers, however, are not professional ethicists, and the disclosure of medical errors is not always a component of their ethical behavior.

In dentistry, medical errors include (a) incorrect medication prescription, (b) neglect of current scientific evidence regarding treatment, (c) improper maintenance of equipment, and failure to (d) properly maintain patient records, (e) acquire informed consent, (f) establish and maintain appropriate infection control measures, (g) accurately diagnose a dental condition, (h) prevent accidents or complications, (i) pursue appropriate follow-up care, and/or (j) follow statutory rules or regulations reflecting quality standards for dental care (Negalberg, 2015). Thusu et al. (2012) showed that the most frequently reported incidents in the practice of dentistry were clerical errors (36%) followed by patient injuries (10%), medical emergencies (6%), accidental ingestion or inhalation of clinical materials (4%), adverse reactions (4%), and erroneous tooth extractions (2%).

Although dentists have an ethical responsibility to fully disclose errors, in practice there is considerable inconsistency regarding opinions on the information that should be disclosed, and who should disclose this information (Blood, 2015). Thusu et al. (2012) reported a relatively low frequency of dental error disclosure, which they attributed to the voluntary nature of reporting and the reluctance of dentists to disclose incidents for fear of loss of earnings.

The disclosure of medical errors varies between clinical specialties (Blood, 2015, Chiodo et al., 1999, Ozar and Sokol, 2002, O’Connor et al., 2010, Yamalik and Perea, 2012). Accordingly, dentists may carry different attitudes than medical doctors toward ethical duty for disclosure. Possible reasons for this discrepancy are hypothesized as follows. First, dental errors may be perceived as less serious. Second, medical care is most frequently provided at large institutions (e.g., hospitals), while dental care is generally more isolated at private practices. Third, medical care is generally provided by a team of doctors, while dentistry is often individually handled. Despite these differences, all medical practitioners, including dentists, have the same ethical obligation to tell the truth, respect patient autonomy, and disclose errors. The disclosure of dental errors is desired by patients and is also recommended by ethicists and professional organizations to ensure that the dental profession can be trusted (Chiodo et al., 1999, Blood, 2015). A critical examination of personal preferences and perceptions of the norm in current practice regarding the disclosure of dental errors is therefore necessary to the benefit of patients, dentists, and the practice of dentistry.

The aim of the current study was to obtain empirical information on the patterns of dental error disclosure among dental professionals in Saudi Arabia. In addition, personal preference, the perception of the norm, and perception of current practices relating to error disclosure were investigated at two levels: the nature (seriousness) of the error that requires disclosure, and the individual responsible for disclosure. To the best of our knowledge, this is the first study to investigate disclosure of errors in the practice of dentistry in Saudi Arabia.

Materials and methods

Study sample

A cross-sectional study was conducted. A questionnaire was administered to dental professional staff and auxiliaries (e.g., dental assistants, dental hygienists, laboratory technicians, and dental nurses) recruited from government and private institutions in Riyadh from August 2015 to January 2016. A power analysis revealed a recommended sample size of 586 based on the prevalence of disclosure of any type of error as 50%, with a 4% margin of error and a confide Web site http://www.raosoft.com/samplesize.html.

Questionnaire

The questionnaire was modified from Hammami et al. (2010) and contained 11 items. Table 1 presents the first five items asking participants to report on which type of dental errors should be disclosed. Responses were classified as “Do not disclose” errors, disclose errors leading to “Major harm”, disclose error leading to “Moderate harm”, disclose errors leading to “Minor harm”, and disclose errors even if “No harm” has been done to the patient. Participants were asked to report on their own personal belief, their perceptions of the norm (i.e., what is appropriate in general/should be done), and their perceptions of current practice. Table 2 presents the second part of the questionnaire, which consisted of six items inviting the participants to report on who (from a list of dental professionals and auxiliaries) should disclose dental errors. Again, participants were asked to report on their own personal belief, perceptions of the norm and perceptions of current practice. Prior to the survey, the modified questionnaire was reviewed by several experts in the fields of ethics and epidemiology, and a pilot study with 10 dentists was conducted to assure content validity.

Table 1

Questionnaire items asking which dental error should be disclosed?

Abbreviation	Which dental error should be disclosed?
Do not disclose	I prefer not to be informed about any dental error that occurred during my dental care
Major harm	I prefer to be informed about a dental error that occurred during my dental care if it caused a major harm (e.g. performing an unnecessary trigeminal nerve damage surgery or failed implants)
Moderate harm	I prefer to be informed about a dental error that occurred during my dental care if it caused at least a moderate harm (e.g. performing an unnecessary vertical root fracture or mechanical pulp exposure or extracting the wrong tooth)
Minor harm	I prefer to be informed about a dental error that occurred during my dental care if it caused any harm, even a minor one (e.g. a soft tissue injury or inadequate crowns or inadequate fillings)
No harm	I prefer to be informed about a dental error that occurred during my dental care even if it did not cause any harm (e.g. a dentist orders the wrong medication but the pharmacist doesn’t dispense it)

Table 2

Questionnaire items asking who should disclose dental errors?

Abbreviation	Who should disclose dental errors?
Dental assistant	A dental assistant in the clinic can inform me about the dental error that occurred to me
Receptionist	A receptionist in the front desk can inform me about the dental error that occurred to me
Erring dentist	I prefer that the dentist who committed the dental error informs me about the dental error that occurred to me
Division head	I prefer that the division head of the dental specialist who committed the dental error informs me about the dental error that occurred to me
Patients relation service	I prefer that the patient‘s relation of the dental service informs me about the dental error that occurred to me
Chairman	I prefer that the chairman of dental department informs me about the dental error that occurred to me

Data collection

Convenience and cluster sampling was used to recruit participants. Twitter, an online social media site, was used for convenience sampling. Cluster sampling was achieved by the recruitment of participants from medical and dental colleges in five regions (Central, North, East, West and South) of Riyadh. A link was posted on Twitter to recruit participants and to advertise the survey. Requests were sent to both individuals and organizations to re-tweet the survey link. A shorter version of the study’s URL was generated to fit within the 140-character limit of Twitter postings. All eligible participants were asked to complete the online questionnaire using Google forms as the platform. This platform facilitated secure, anonymous data collection and ensured confidentiality. Exclusion of duplicate data was conducted by reviewing the IP addresses of the respondents.

Statistical analysis

Statistical analyses were conducted using SPSS version 20. The demographic characteristics of participants were summarized. Two cross-tabulations were constructed from the questionnaire responses. The first cross-tabulation summarized the frequencies of participants who reported which dental error to disclose (in the rows) depending on personal belief and perception of the norm of current practice (in the columns). The second cross-tabulation summarized the frequencies of responses to who should disclose dental errors (in the rows) depending on personal belief and perceptions of the norm of current practice (in the columns).

Because each participant chose only one item from the list in Table 1 and one item from the list in Table 2 the frequency distributions of the choices in the cross tabulations represent hierarchical rankings of the items, and therefore the variables used in the statistical analysis were measured at the ordinal level. Kendall’s coefficient of concordance for ranks was used to determine the extent of agreement between the three dimensions of personal belief, perception of the norm and perception of practice with respect to which type of error should be disclosed. Wilcoxon’s signed rank tests were used to conduct pairwise comparisons between the related measures of which type of error should be disclosed across the three dimensions (i.e., belief vs. norm; belief vs. practice; and norm vs. practice).

The demographic data violated the theoretic assumption of an ordinal logistic regression (Hosmer and Lemeshow, 2000) and therefore, the types of errors that should be disclosed were collapsed into a binary format, and a binary logistic regression analysis was conducted. The forward stepwise procedure was applied to select the predictors, so that only the statistically significant demographic categories (indicated by p < 0.05 for the Ward test statistic) were included in the models. Demographic categories that were not significant predictors (p ≥ 0.05) were excluded. Hosmer and Lemeshow goodness of fit tests confirmed that the data significantly fitted the logit function at the 0.05 level.

It is imperative to report and explain the coding of all variables to reliably interpret the results of a logistic regression (Bagley et al., 2001). For the first analysis, the dependent variable was coded as 1 = “Disclose” (i.e., all choices, except “Do not disclose”) or 0 = “Do not disclose”. For the second analysis, the dependent variable was coded as 1 = error resulting in “No harm” or 0 = error resulting in “Harm” (i.e., all choices except “No harm”). To ensure that the sample size in each of the demographic categories was large enough to provide sufficient statistical power to construct accurate models without Type II errors (Bagley et al., 2001), the categories for the institutions and occupations were collapsed and coded as follows. Institution: 1 = Government, 0 = Private; Occupation: 1 = Dental Specialist, 0 = Other occupations; Age (Years): 1 = <30, 2 = 31–40, 3 = >40. The other demographic categories were coded as follows: Nationality: Saudi = 1, Non-Saudi = 0 and Gender: Male = 1, Female = 0.

Results

Demographic characteristics of participants

The demographic characteristics of the participants are summarized in Table 3. The sample size was N = 586 with approximately equal proportions of male and female participants. The most frequent age group was <30 years (n = 364, 62.1%) and the least frequent was >50 years (n = 27, 4.6%). The majority of participants were Saudi (n = 427, 72.9%). Most participants were recruited from six government institutions (n = 353, 60.3%) with the remainder from more than three private colleges or clinics. The most frequent occupations were dental specialist (n = 148, 25.3%) and student (n = 204, 34.8%) with smaller proportions of residents, academics, and dental auxiliaries (e.g., dental assistants, dental hygienists, laboratory technicians, and dental nurses).

Table 3

Demographic characteristics of participants.

Characteristic	Category	n	%
Gender	Male	297	50.7
	Female	289	49.3
Age (Years)	<30	364	62.1
	30–39	139	23.7
	40–49	56	9.6
	≥50	27	4.6
Nationality	Saudi	427	72.9
	Non-Saudi	159	27.1
Government Institutions	King Abdulaziz Medical City	131	22.4
	King Saud University	57	9.7
	Princess Noura University	38	6.5
	Other government medical cities	48	8.2
Private Institutions	Riyadh Colleges of Dentistry & Pharmacy	172	29.4
	Al Farabi Colleges	32	5.5
	Other private clinics	29	4.9
Occupation	Dental Specialist	148	25.3
	Undergraduate Student	204	34.8
	Resident	112	19.1
	Academic	68	11.6
	Dental auxiliary	54	9.2

Which type of dental errors should be disclosed?

Table 4 presents the cross-tabulation of the frequencies and percentages of the 586 participants endorsing which type of dental errors should be disclosed depending on the participant’s personal belief, perceptions of the norm and current practice. Very few participants reported a personal belief of not disclosing dental errors (5.6%) implying that 94.4% preferred disclosure. Slightly higher proportions or participants perceived that not disclosing errors was the norm (5.8%) or was currently practiced (9.6%). Relatively few participants reported that disclosing “Major harm” was their preference (9.6%); or was a perceived norm (8.4%) or was currently practiced (17.9%). The most frequent responses were (a) 49.0% reported that their personal belief was to disclose errors resulting in “No harm”; (b) 45.4% reported that the norm was to disclose errors resulting in “Minor harm”; and (c) 32.3% reported that the current practice was to disclose errors resulting in “Minor harm”.

Table 4

Cross-tabulation of the frequencies (n) and percentages (%) of participants reporting which type of dental error should be disclosed, depending on the respondent’s personal preference and perceptions of norm and current practice.

Which type of dental error should be disclosed?	Preference n (%)	Norm n (%)	Practice n (%)
Do not disclose	33 (5.6)	34 (5.8)	56 (9.6)
Major harm	56 (9.6)	49 (8.4)	105 (17.9)
Moderate harm	41 (7.0)	69 (11.8)	129 (22.0)
Minor harm	169 (28.8)	266 (45.4)	189 (32.3)
Near miss	287 (49.0)	168 (28.7)	107 (18.3)
Total	586 (100)	586 (100)	586 (100)

Kendall’s W = 0.173, p < 0.001. Pairwise comparisons using Wilcoxon’s signed rank test for Preference vs. Norm: p < 0.001; Preference vs. Practice: p < 0.001; Norm vs. Practice: p < 0.001.

Kendall’s coefficient evaluated participant agreement across the three dimensions of disclosure (belief, norm, and practice) for each type of dental error, based on a scale from 0 (no agreement) to 1 (100% agreement). While Kendall’s coefficient was statistically significant (p < 0.001) the low W value (0.173) reflected a weak level of agreement. Pairwise comparisons using Wilcoxon’s signed rank tests reflected significant (p < 0.001) differences between the ranked responses for belief vs. norm, preference vs. practice, and norm vs. practice.

Who should disclose dental errors?

Table 5 presents the cross-tabulation of the frequencies and percentages of the 586 participants endorsing the person responsible for disclosure of dental errors, depending on personal belief, perceptions of the norm and perceptions of current practices. Most participants chose the erring dentist as the persons who are responsible for disclosing error as their personal belief (68.9%), as the perceived norm (73.2%) and as the current practice (68.4%). Disclosure of dental errors by the division head, the dental assistant, the patient’s relation service, and the chairman was less frequently reported as a personal belief, norm, or practice. The least frequent response was for the dental error to be disclosed by the receptionist (personal belief: 2.0%; perceived norm: 3.6%; and the current practice: 3.9%).

Table 5

Cross-tabulation of the frequencies (n) and percentages (%) of participants reporting who should disclose dental errors, depending on the participants’ personal preference and perceptions of norm and current practice.

Who should disclose dental errors?	Preference n (%)	Norm n (%)	Practice n (%)
Dental assistant	52 (8.9)	39 (6.7)	41 (7.0)
Receptionist	12 (2.0)	21 (3.6)	23 (3.9)
Erred dentist	404 (68.9)	429 (73.2)	401 (68.4)
Division head	59 (10.1)	46 (7.8)	56 (9.6)
Patient’s relation service	31 (5.3)	29 (4.9)	41 (7.0)
Chairman	28 (4.8)	22 (3.8)	24 (4.1)
Total	586 (100)	586 (100)	586 (100)

Kendall’s W = 0.366, p < 0.001. Pairwise comparisons using Wilcoxon’s signed rank test for Preference vs. Norm: p = 0.674; Preference vs. Practice: p = 0.381; Norm vs. Practice: p = 0.134.

Kendall’s coefficient evaluated the degree to which participant responses agreed among the three dimensions of disclosure (belief, norm, and practice) for each category of person responsible for disclosure. The Kendall’s coefficient was statistically significant (p < 0.001) but the W value (0.366) reflected only a moderate level of agreement. Pairwise comparisons using Wilcoxon’s signed rank tests reflected no significant (p > 0.05) differences between the ranked responses for preference vs. norm, preference vs. practice, and norm vs. practice. In conclusion, the participants generally agreed with who should disclose dental errors. The consensus, based on the consistently high levels of endorsement in Table 5, was that disclosure was mainly the responsibility of the erring dentist.

Prediction of the likelihood of disclosing dental errors

Table 6 presents the three binary logistic regression models to predict the likelihood of disclosing dental errors according to the resulting patient harm (i.e., “Major harm”, “Moderate harm”, “Minor harm”, or “No harm”) vs. “Do not disclose” as the reference, with respect to the three dimensions of disclosure. The preference for disclosing dental errors tended to be less frequent among older compared to younger participants. The odds ratio indicated the likelihood of preferring to disclose errors decreased by a factor of 0.48 (95% CI = 0.30, 0.77) for every one unit increase in the ordinal age scale (i.e., between 1 = <30 and 3 = ≥40). In addition, the preference for disclosing errors was greater for participants at government institutions compared to participants at private institutions. The odds ratio indicated the likelihood of a participant at a government institution disclosing errors was 6.32 (95% CI = 2.73, 14.66) times greater than a participant at a private institution.

Table 6

Binary logistic regression to predict disclosure of dental errors vs. other choices.a

Dimension	Significant predictor	p	Odds ratio	95% CI
Preference	Age	0.002	0.48	0.30	0.77
	Institution	<0.001	6.32	2.73	14.66
Norm	Institution	<0.001	6.48	2.77	15.14
Practice	Institution	<0.001	2.94	1.62	5.32
	Nationality	0.028	1.92	1.07	3.42

Dependent variable: 1 = Disclose; 0 = Do not disclose (all other choices in Table 1).

Participants reported that the perceived norm was for government institutions to be more likely to disclose dental errors than those at private institutions. The odds ratio indicated that the perceived norm was for participants at government institutions to be 6.48 times (95% CI = 2.77, 15.14) more likely to disclose dental errors than those at private institutions. Likewise, participants at government institutions were perceived to be more likely to disclose dental errors than those at private institutions. The odds ratio indicated that participants perceived government institutions to be 2.94 times (95% CI = 1.62, 5.32) more likely to disclose dental errors than those at private institutions. Lastly, participants perceived that Saudis were 1.92 times more likely to disclose dental errors compared to non-Saudis (95% CI = 1.07, 3.42).

Prediction of the likelihood of disclosing errors resulting in “No Harm”

Table 7 presents the three binary logistic regression models using demographic variables to predict the likelihood of disclosing errors resulting in “No harm” vs. Harm (i.e., “Major harm”, “Moderate harm”, “Minor harm”) or not disclosing errors at all (“Do not disclose”), with respect to the three dimensions of disclosure. The odds ratio indicated that the likelihood of preferring to disclose errors resulting in “No harm” was 0.51 times less (95% CI = 0.37, 0.72) in males than in females. Furthermore, the preference for disclosing errors resulting in “No harm” was 2.36 (95% CI = 1.67, 3.35) times greater for participants at government institutions compared to those at private institutions.

Table 7

Binary logistic regression to predict disclosure of “No harm” vs. other choices.a

Dimension	Significant predictor	p	Odds ratio	95% CI
Preference	Gender	<0.001	0.51	0.37	0.72
	Institution	<0.001	2.36	1.67	3.35
Norm	Institution	0.037	1.50	1.02	2.19
	Age	0.012	0.72	0.55	0.93
Practice	Institution	0.011	1.83	1.15	2.92

Dependent variable: 1 = “No harm”; 0 = All other choices in Table 1.

The odds ratio indicated that participants at government institutions were 1.50 times more likely (95% CI = 1.02, 2.19) than participants at private institutions to disclose errors resulting in “No harm” as the norm. The odds ratios also indicated the likelihood of preferring to disclose errors decreased by a factor of 0.72 (95% CI = 0.55, 0.93) for every one unit increase in the ordinal age scale. Consequently, older participants were less likely to disclose errors resulting in “No harm” than younger participants. The perceived practice was for participants at government institutions to be more likely to disclose errors resulting in “No harm”. The odds ratio indicated that participants perceived that government institutions in practice were 1.83 times (95% CI = 1.15, 2.92) more likely than those at private institutions to disclose “No harm”.

Discussion

Empirical data were obtained from 586 participants at over 10 dental institutions in Riyadh, Saudi Arabia, regarding issues related to the disclosure of dental errors. The research focused on the associations between the professional responsibility for disclosure (categorized by personal belief, perception of norm, and perception of current practice) and the error factors (specifically the nature of the dental error to be disclosed) and the institutional culture (specifically who should disclose the error). Statistical evidence based on the analysis of cross-tabulated data was consistent with the conceptual model of Fein et al. (2005) positing that three of the most important influences controlling whether to disclose medical errors involved provider factors, error factors, and institutional culture.

The current findings revealed considerable differences of opinion among participants regarding the level of dental error disclosure. The personal beliefs of participants, their perceptions of the norm and the current practices with respect to which type of dental errors should be disclosed were inconsistent, and the level of agreement between the three dimensions of disclosure was weak. These finding are in agreement with those of Blood (2015) who reported considerable inconsistency among dentists regarding how much information they disclosed about their mistakes.

The finding that 94.4% of the participants in the current survey preferred that dental errors be disclosed was consistent with the high proportions of participants in previous studies, including both patients and providers, who have endorsed the disclosure of medical errors (Hammami et al., 2010). Despite the high level of preference for disclosing errors, the percentage of participants that perceived that it is current practice to disclose errors was relatively low (17.9%, 22.0%, 32.3% and 18.3% to disclose errors resulting in “Major harm”, “Moderate harm”, “Minor harm” and “No harm” respectively). These results imply a perception that there is extensive non-reporting of dental errors and are in agreement with other studies, suggesting that less than half of medical errors may be disclosed to patients (Blendon et al., 2003, Hammami et al., 2010). It appears that although healthcare providers are ethically bound to admit mistakes to patients, in reality, most practitioners and institutions do not disclose errors.

Hammami et al. (2010) found a similar distribution of responses for perception of the norm and personal beliefs regarding type of medical errors that should be disclosed among patients attending outpatient clinics in Saudi Arabia, suggesting that the personal beliefs of the Saudi culture tended toward the norm. In the current study, however, the level of agreement between personal beliefs and the norm was relatively weak, implying a possible difference in attitudes between dentists and medical doctors.

Although there were inconsistencies as to who should disclose dental errors, the majority (over 68%) of the participants reported a personal belief, a perception of the norm, and perception of practice, that medical errors should be disclosed by the erring dentist, with a moderate level of agreement. Consequently, the level of agreement between personal belief, perceived norm, and perceived practice was stronger for the person responsible for reporting errors compared to which errors should be disclosed.

The current study found that the likelihood of disclosing dental errors differed between participants at government and private institutions. It is possible that participants at government institutions are more likely to report dental errors, including errors leading to “No harm” to the patient as a personal belief, perception of the norm, and perceived practice, because they feel more accountable to the public for their actions, compared to those at private institutions. Other significant demographic predictors were that (a) males were less likely to personally prefer the disclosure of errors leading to “No harm” than females, (b) older participants were less likely than younger participants to prefer disclosure of dental errors as well as disclosure of errors resulting in “No harm” as the norm, and (c) there was a perception that Saudis were more likely than Non-Saudis to disclose dental errors in practice. Hammami et al. (2010) similarly found age and gender to be predictors of disclosure of medical errors in Saudi Arabia. Older age and female gender predicted a preference for disclosure of errors leading to “No harm” while younger age and male gender predicted a preference not to disclose errors.

Healthcare organizations in several countries have developed guidelines to help providers adopt an ethical course of action for the disclosure of medical errors (Chafe et al., 2009, Williams, 2012). However, according to Hammami et al. (2010) the Implementation for Regulation of the Practice of Medicine and Dentistry released by the Saudi Ministry of Health in 1990 and the Ethics of Medical Profession released by the Saudi Commission of Health Specialists are “silent on this issue”. Different cultures may require different levels of disclosure and the needs and demands related to disclosure of medical errors may be different in Saudi Arabia compared to those in other countries. The findings from the present study can assist the Ministry of Health in the development of new ethical policies for the disclosure of dental errors in Saudi Arabia. Further research regarding the attitudes and practices of dentists toward the disclosure of dental errors in the context of the Islamic/Arabic Culture will also be necessary to complement the current findings.

Ethical statement

This survey was conducted in compliance with ICH-GCP Ethical Standards and Research Protocol #RSS 15/045 approved by Institution Review Board (IRB) of King Abdullah International Medical Research Center (KAIMRC) in the period from August 2015 to January 2016. All participants provided verbal consent.

Conflict of interests

The authors have no known conflict of interests associated with this study that could have influenced its outcome.