Physicians’ Perception and Practice of Prescribing Vitamin B Combination Versus Antiepileptic Drugs for Diabetic Neuropathy: Content Validity, Reliability and Pilot Study

Diabetic neuropathy (DN) is the most common complication of diabetes mellitus causing increased morbidity and mortality. Although international guidelines did not include vitamins and dietary supplements in any line of management, these agents were prescribed by a significant number of physicians either as preventive or treatment of DN. This study aimed to develop a validated questionnaire that examines the physicians’ perception and practice towards prescribing vitamins or dietary supplements for DN treatment. The questionnaire was developed upon literature review via PubMed, Medline, Web of Science, Scopus, and Google Scholar. The questionnaire was revised upon experts’ views and opinions. The constructed questionnaire was validated by the means of content validity and internal consistency. Finally, the developed questionnaire was piloted in a small representative sample of the original future aimed research target population, to test its applicability and feasibility. A total of 22 items questionnaire were developed in two domains (knowledge and perception domain and practice domain). Content validity analysis results met a satisfactory level, in which the S-CVI/UA value was 0.86 for both clarity and relevance. The questionnaire also showed good reliability, in which the Cronbach’s alphas was 0.805. In conclusion, this study showed the that constructed questionnaire had a good level of validity (content validity) and reliability that is able to cover different aspects of the current state of perception and practice among physicians regarding the management of DN. The preliminary results of the pilot study showed a good knowledge and perception of the respondents with DN. Even in the absence of a local guideline, the respondents followed the international guidelines in choosing anti-neuropathic agents however, they tended to add vitamin B complex/B 12 as an adjuvant in their management plan.


Introduction
Diabetes mellitus is defined as a metabolic disorder caused by multifactorial aetiologies and presented by chronic hyperglycemia with defects in carbohydrate, fat, and protein metabolism resulted from either abnormal insulin secretion, insulin action or both (1) . In 2021, the International Diabetes Federation stated that there are about 537 million diabetes cases, and this number is projected to reach 643 million by 2030, and 783 million by 2045 (2) . The rapid escalation in the number of cases will be accompanied by a direct increase in the prevalence of the chronic complications of the disease (3) . Diabetic neuropathy (DN) is the most common complication of diabetes mellitus (4) causing increased morbidity and mortality (5) furthermore causing a great burden on health care expenditure (6) . DN is defined as a neurodegenerative disorder that affects the somatic and/ or autonomic peripheral nervous system in the setting of diabetes mellitus without other causes of neuropathy (7) . Despite its heavy burden on healthcare systems and the better understanding of the multifactorial pathogenesis of the disease, DN still underdiagnosed and undertreated (8) . In the absence of evidence-based disease-modifying pharmacological agents, the mainstay of the management is based on glucose control, lifestyle changes, and pain management (9) . Despite the relatively high prevalence of DN in Iraq (10) , there is no local or regional guideline of DN management (11) . Neurotropic B vitamins (B1, B6 and B12) were well-known to have a crucial role in maintaining healthy nervous system (12) . It was suggested that these vitamins may increase the availability and effectiveness of noradrenaline and 5-hydroxytryptamine in the descending inhibitory nociceptive system (13,14) furthermore, in some countries they were classified as analegesic drugs (14) .It was hypothesized that neurotropic B vitamins support directing the Wallerian degeneration process (which is an active process that commenced after a nerve injury of degeneration of an axon) toward regeneration and remyelination (15) , several clinical trials conducted to highlight the effect of administration of the neurotropic B vitamins together with different dose regimes (16 -18) , however the positive outcomes was limited to improvement in symptoms of pain numbness and paraesthesia with no improvement in neurophysiological parameters. Thiamin (B1) had been shown to serve as a cofactor for three important enzymes that were involved in carbohydrate metabolism (19) , in addition to have an important role in abolishing the metabolic dysregulations caused by high glucose level including polyol pathway, protein kinase C pathway, hexosamine pathway and increased advanced glycation end products which are the proposed mechanisms of DN pathogenesis (19,20) . Benfotiamin which is a synthetic derivative of thiamin with better pharmacokinetic properties (21,22) was shown to improve the Neuropathy Symptom Score in a double-blind randomized placebocontrolled phase III clinical study (BENDIP) (23) . Despite the fact that the association between vitamin B6 level and diabetes mellitus was well established (24) , two clinical trials of pyridoxin administration in patients with DN were failed to improve subjective and objective parameters of the disease (25,26) . Furthermore, concerns were raised about the safety profile of pyridoxin following several case reports about its possible toxic effects on nerves (27,28) which was proposed to be due to disruption of Gamma-Aminobutyric Acid (GABA) biosynthesis (29) which led several Food Safety Authorities is several countries to decrease the allowed upper limit of pyridoxin dose (30,31) . A long-term trial was showed a significant effect of metformin on B12 and methylmalonic acid (MMA) level with concurrent worsening of neuropathy symptoms (32) . Several clinical trials concluded a positive effect of methylcobalamin (active form of vitamin B12) on the symptoms of DN with no effects on the nerve study parameters (33 -35) until recently (2021) a prospective one-year, randomized double-blind, placebocontrolled trial, concluded positive effects of 1 mg methylcobalamin on in sural nerve conduction parameters along with Michigan Neuropathy Screening Instrument Questionnaire (MNSIq), the level of pain, and the quality of life (36) . Although the latest international guidelines did not include vitamins and dietary supplements in any line of management (e.g., the National Institute for Health and Care Excellence National (NICE) updated guideline in 2020 (37) and the American Diabetes Association (ADA) guideline in 2021 (38) ). A recent study, that was conducted in Saudi Arabia, found that a significant number of physicians were prescribing vitamin B12 either as preventive or treatment to their patients with DN (55) . The supplements have a heavy economic burden for patients in a developing country like Iraq, which lacks national health insurance programs. In addition to the fact that the safety profile of some heavily prescribed vitamins is questionable (39) . This study aimed to develop a validated questionnaire to assess physicians' perception and practice towards the prescribing of vitamins and dietary supplements versus pharmacological agents for the management of DN.

Methods
The study design involves three phases. In phase one, the construction of the first draft of the questionnaire from an interpretation of the literature on the research topic. In phase two, the content validity of the drafted questionnaire was tested to verify the applicability and appropriateness. Finally, in phase three, the internal consistency of the questionnaires was piloted to ensure reliability and feasibility.

Phase one
The literature review was conducted between the 1 st of October 2021 and the 30 th of December 2021 for published articles before the time of the research (no time limit was set to get as many as possible articles on the research topic). The search was only for full articles, written in English or Arabic languages that are examining healthcare professionals' or physicians' knowledge, attitude, perception, and practice toward neuropathic pain treatment or management. The exclusion criteria were; articles written in languages other than English or Arabic, abstracts, articles examining patients' perspectives rather than physicians. The literature review was conducted by the first author, a master student and the chief investigator, which were trained to search for relevant articles via the available databases and search engines including PubMed, Medline, Web of Science, Scopus, and Google Scholar. The Medical subject headings (MeSH) terms used in the search were supplied as a supplementary file. The number of results retrieved from the literature review through the above-mentioned databases was 86 researches of a variety of formats. Following a screening of the retrieved researches and excluding irrelevant ones, the final number of researches chosen was 10 ( Figure 1).

Figure 1. Workflow showing the process of literature review for constructing the first draft of the questionnaire
Based on the results of the literature review, the ten chosen articles were reviewed and used in the construction of the initial draft of the questionnaire. The questionnaire was constructed in English language since Iraqi physicians are more familiar with medical terminology in English rather than Arabic language. Then the questionnaire was sent independently to two experts in KAP (knowledge, attitude, and practice) tools at the University of Baghdad and the University of Duhok to verify the relevance and appropriateness of the questionnaires for the study topic. Based on the experts' views, the first draft was amended accordingly to improve the clarity and the structure of the questionnaire. This was followed by a second review process by another independent reviewer at the college of Pharmacy, University of Mosul to evaluate the questionnaire and eliminate redundancy and ambiguity, and to develop a final draft of the questionnaire. The questionnaire was developed to explore the physicians' perception and practice towards the prescribing of vitamins and dietary supplements versus pharmacological agents for the management of DN. Therefore, the items (questions) were allocated into two main domains (knowledge and perception domain and practice domain). The knowledge and perception domain aimed to closely examine the basic knowledge and the general perception of the physicians about DN (including the commonality of DN, its effect on patient's quality of life, mortality rate and the difficulty of treatment in addition to the perception about dietary supplements and vitamin B). The practice domain aimed to explore the physicians' prescribing plans regarding anti-neuropathic agents and dietary supplements and indifferent cases. In addition to the factors that affect the choice of these agents. The time expected for completing the questionnaire (10 -20 minutes) was enough for almost all of the respondents.

Content validity
The six steps of the content validity procedure were conducted as follow: (I) content validity form development, (II) assigning an expert review panel, (III) distributing the content validation form to the expert (d) reviewing the domain and items (e) setting a score to each item (f) determining the content validity index (CVI) (40). In which the CVI uses item-CVI (I-CVI) and scale-CVI (S-CVI) to report content validity in questionnaire development (41). The I-CVI is determined by dividing the total number of experts by the number of experts who rated each item on a scale of 'very relevant' (score 1), 'relevant' (score 1), 'irrelevant' (score 0), and 'very irrelevant' (scoring 0) (42). The I-CVI value range of 0.8 or more indicates that the item is relevant, however, if the value is less than 0.8 and more than 0.69, this would suggest item revision, while a value of less than 0.7 indicates an item to be eliminated. The sum of all items with I-CVI equal to 1 divided by the total number of items for the S-CVI yielded the universal agreement (UA) among experts (S-CVI/UA). A content validity index of S-CVI/UA 0.8 indicates outstanding content validity (43). Eight experts, including physicians and pharmacists from the University of Mosul and Ninevah University, as well as retired physicians, were recruited to judge the items using a scoring system ranging from 1 (the item is not relevant at all) to 4 (the item is highly relevant). If the items were scored lower than 3, in the comments, the experts were asked to offer a note or a suggestion to rephrase the item.

Reliability
Internal consistency is an indicator of a questionnaire's homogeneity (44). It reflects how close the questions (items) of a questionnaire are intercorrelated and hence measure the same concept. Cronbach's alpha (α) value is used as a measure of internal consistency (45). The internal consistency reliability of the two domains (knowledge and perception domain, and practice domain) was tested using Cronbach's alpha coefficient. Forty-five participants were enrolled to complete the questionnaire, the sample size was satisfying the Central Limit Theorem which assumed that the distributional assumption of the sample size of 30 or more to guarantee an equal mean between any sample and the target population (46).

Pilot study
To test the feasibility and usability of the questionnaire, a pilot study was conducted. A total of 45 physicians in different specialties were recruited. A convenience purposive sampling from Avicenna Hospital and Al-Quds primary health centre in Mosul, Iraq. Ethical approval was obtained from the Medical Research and Ethics Committee of the Department of clinical pharmacy, College of Pharmacy, University of Mosul. Ethical approval was also obtained from the Collegiate Committee for Medical Research Ethics at the University of Mosul (code: CCMRE-phA-22-2) to conduct the study. The physicians were contacted to participate in this pilot via telephone and verbal consent was given to participate in the study. On the day of the interview, written consent was given by the participants. The researcher discussed the aims of the research and handed the paper survey questionnaire. The questionnaires were filled out in 15 minutes (±3 minutes). The face-to-face data collection started from the 20th to the 31st of January 2022.

Statistical analysis
Data obtained from the respondents (for content validity, reliability and pilot study) were entered to Microsoft Excel (2016) software to calculate the content validity. Then the data was moved to the Statistical Package for the Social Sciences (SPSS) software version 24 to run the reliability test. Descriptive analyses (percentages and frequencies) were carried out using SPSS software for the pilot study.

Phase one
The process of reviewing the literature via PubMed, Medline, Web of Science, Scopus, and Google Scholar databases resulted in 86 researches of a variety of formats (abstracts, posters, full articles, and books). These materials were screened by means of the titles and abstracts to remove duplicates and irrelevant articles. Ten full-text articles were retrieved for further evaluation based on the inclusion and exclusion criteria ( Figure 1). The final number of researches chosen was 10 ( Table 1). The questionnaire was developed from the tools of the retrieved articles with modification to meet the aims of the developed questionnaire and to make it more suitable for the Iraqi healthcare system, which differs massively from other healthcare systems where other articles were conducted. The questions were allocated into two main domains, namely, the knowledge and perception domain, including eight items, and the practice domain including 16 items. The first draft of the questionnaire was reviewed by two experts; these include deletion of two questions for redundancy and irrelevance. The experts also requested to rephrase three more questions to improve the clarity of the questions. On the second round of review, an expert suggested adding a "not sure" for questions with an answer of "yes" or "no" for closed-ended questions. The final draft consisted of 22 items (seven items in knowledge and perception domain and fifteen items in the practice domain) ( Table 2). The effectiveness of an agent is an important factor in choosing DN medication Agree, neutral, disagree

Item 5 The safety of an agent is an important factor in choosing DN medication
Agree, neutral, disagree

Item 6 Patient preference is an important factor in choosing DN medication
Agree, neutral, disagree

Item 7 Drug -drug interaction is an important factor in choosing DN medication
Agree, neutral, disagree

Item 8 Cost of an agent is an important factor in choosing DN medication
Agree, neutral, disagree

Phase two (content validity)
To determine the content validity of the preliminary version of the questionnaire, an expert panel of eight healthcare professionals was recruited. The experts rated the items' (questions') clarity and relevance on a scale from (1 -4) to calculate the I-CVI, S-CVI, and S-CVI/UA. Tables  2 and 3 show the calculation of the validity indices for both clarity and relevance. The validity indices met satisfactory levels (Tables 3, 4 and 5).

Pilot study
A total of 45 physicians (56% male and 44% female) were recruited in piloting the final draft of the questionnaire. The age group of the majority (64%) of the participants was between 35-54 years. They were from different professional levels ranging from consultants (11%), specialists (51%), residents (22%), and general practitioners (16%). They were from different specialties (family medicine 29%, internal medicine 24%, general surgery 11%, and other specialties 36%). About half (49%) of the respondents were working in governmental hospitals while 36% were in primary care centers. Only 42% of them were working in private clinics. More than half (56%) of the respondents have more than 10 years of experience in their field of specialty (Table 7).

Knowledge and Perception domain
In the knowledge and perception domain, the majority of the respondents presented good knowledge of DN in terms of its commonality and effect on patients' quality of life. While only about half of the respondents had good knowledge on the management plan's mainstay and its difficulty along with the effect of DN on mortality rate. In the context of the effect of blood glucose control on DN, the majority of the respondents thought that controlling blood glucose could decrease the incidence of DN while only about one-quarter of the physicians thought that blood glucose control cannot reverse DN (Table 8). Regarding the knowledge of physicians towards vitamins and dietary supplements, about one half of the respondents thought that the dietary supplements are approved by the American Food and Drug Administration (FDA) and that they are quality controlled just like chemical medicines. While about one half of the respondents disagreed with the idea of vitamins and dietary supplements' safety on long-term use, the lack of associated side effects and their lack of tendency to interact with other medications. The respondents were evenly split between supporters and opponents regarding the cost-benefit of vitamins and dietary supplements. About one-half of the respondents believed that vitamin B complex/ B12 has no role in the treatment of DN, nor in the prevention of DN. However, near half of the respondents believed that vitamin B complex/ B12 supplements are safe for long-term use in the treatment of DN. One-half of the respondents stated that no local (Iraqi) guideline for the management of DN is available and about 34% of them were not sure if such guideline exists. Similarly, about twothirds of the respondents thought that there was no definite treatment for DN while about one-third were not sure about the availability of such treatment in the current time. More than one-half of the respondents believed that vascular damage, toxic effects of glucose, and vitamin B12 deficiency are the most common causes of DN. Regarding the source of information for the management of DN, respondents rated textbooks and scientific journals, and international guidelines as the most important sources of knowledge (Table 8).

Practice domain
The majority (80%) of respondents expressed their willingness to prescribe vitamins and/or dietary supplements to their patients with DN in a frequent manner. Vitamin B complex was the most commonly mentioned dietary supplement when they were asked to give an example of dietary supplement that they recommend for their patients. And they recommend these dietary supplements (vitamin B complex) as adjuvant therapy. More than one-half of the respondents considered the effectiveness, safety, cost and patient factors as very important factors in choosing medications for the treatment of DN. More than two-thirds of the respondents chose one of the first line antineuropathic agents according to the latest guidelines for DN management, in which, gabapentin was the most chosen agents for the treatment of DN.The respondents' opinions regarding the influence of international guidelines on their practice were in between consulting the guideline and utilizing the guideline. Regarding their management plan of different degrees of DN, more than two-thirds of the respondents recommended the following lines of treatment: lifestyle change, strict control of blood sugar and dietary supplements (including vitamins) for the different severities of the disease. More than half of the respondents recommended an anticonvulsant as an option in managing moderate and severe DN, while about half of the respondents stated their recommendation of combining anticonvulsant, tricyclic antidepressant (TCA), a dietary supplement with or without serotoninnorepinephrine reuptake inhibitor (SNRI) in managing severe DN (Table 9).

Discussion
The literature review process revealed no validated questionnaire that is targeted to explore physicians' perception and practice towards the prescription of vitamins and other medicines in DN management. However, the review process found two articles that explored the utilization patterns of anti-neuropathic i.e (TCA, SNRI, anticonvulsants, and opioids) in the UK and Korea (47,56) . The prescribing preferences of healthcare professionals (physicians, pharmacists, and nurses) regarding antineuropathic medications, their knowledge, and beliefs about DN were examined in six researches (48)(49)(50)(51)(52)(53) . While Provenzano et al., (2018) have investigated the practice of primary care physicians about the management of chronic pain (54) . Aldossari et al., (2021) had assessed the physicians' knowledge and the tendency of prescribing vitamin B12 exclusively for the treatment or prevention of DN in primary care settings (55) . Our study illustrated the ability to produce a questionnaire with good psychometric properties on the physicians' perception and practice toward prescribing vitamin B combination versus antiepileptic drugs in the management of DN along with the factors affecting their choice and the influence of the international guidelines to their current practice. The questionnaire of this study was developed by adapting items from the former-mentioned studies in addition to the self-constructed questions. The content validity was measured in this study since it is an essential step in questionnaire's development. Furthermore, it is an acceptable and important method to reflect the relevance and clarity of the instrument (57) . The content validity indices ranged from 0.8 to 0.9 for both relevance and clarity which is satisfactory and reveals that the questions were relevant and representative of the intended aim of the developed tool (58) . Shi et al., (2012) indicated that I-CVIs ≥ 0.78, S-CVI/UA, and S-CVI/Ave ≥ 0.8 are considered excellent indices for relevance and clarity. The content validity indices of the constructed questionnaire of this study were within the above-mentioned ranges (59) . The use of Cronbach's alpha in the multiple-item questionnaire was considered as a routine in knowledge, attitude, and practice studies (60) , which is used to test and determine the internal consistency (reliability) of a questionnaire in applied research (61) . Nunnally and Bernstein (1978) hypothesized that an alpha value of 0.70 and higher is an acceptable value in the early stages of developing a questionnaire (i.e., exploratory research) (62) . On the other hand, George and Mallery (2003) ranked alpha values according to acceptability into excellent ≥ 0.9, good ≥ 0.8, acceptable ≥ 0.7, questionable ≥ 0.6, poor ≥ 0.5, and unacceptable ≤ 0.5. The Cronbach's alpha coefficient for internal consistency of this study was 0.804 which is considered good (63) . Although the pilot study is time-consuming and leads to considerable data loss, it ensures the feasibility of the developed questionnaire, in other words, it detects any inappropriate and/or complicated questions that would ultimately end up by potential failure of the questionnaire. In our pilot study, the clarity of the questions was confirmed by face-to-face short interviews with the respondents from different specialties (family medicine, internal medicine, general surgery, endocrinology, rheumatology, and neurology). As expected, the physicians showed a good perception of the commonality of DN, the impact of the disease on quality of life, and the role of glucose control in the etiology of DN and the management plan of the disease. Diverse opinions were noticed regarding the process of manufacturing, FDA approval, safety profile, and the cost-effectiveness of vitamins and dietary supplements. The lack of a local guideline was raised by the respondents as a reason for their different perspectives regarding the use of vitamins and medicines for the management of DN, in which their main source of information were reputable international guidelines like the American diabetes association (ADA) and NICE along with textbooks and scientific journals. However, physicians opted gabapentin, pregabalin, and duloxetine as their first pharmacological choice which was in line with the current international guidelines, the percentage of Iraqi physicians who were adhered to international guidelines was higher than the percentage of the recent Saudi study (55) . Although we expected that physicians would prioritize the efficacy and safety in their treatment choice, they also prioritized cost as an important factor in choosing the medication. Interestingly, physicians are willing to prescribe vitamins and supplements for therapeutic purposes as an adjuvant which was consistent with the results of the Saudi study (55) in which about half of the physicians rated the efficacy of vitamin B12 as moderate in both prevention and treatment of DN. Regarding the management of mild, moderate, and severe DN, although the physicians adhered to the international guidelines that recommended lifestyle changes and strict blood glucose control as a primary intervention for DN. However, they unfollowed these guidelines by recommending vitamins and dietary supplements as an adjuvant treatment. This study has some limitations; first is that despite the fact that Cronbach's alpha is a commonly used test to determine the internal consistency in the literature it has limitations when used as a sole index for internal consistency. Second, the small number of participants in the pilot study was the consequence of the fact that the target population (physicians) are limited in number and hard to reach, and that those who are participated in the pilot study would not be included in the main study (since this article is a part of a larger study and Master degree dissertation); therefore, losing this number would affect the power of the main study. To the best of our knowledge, this study is the first to develop and validate a tool to explore physicians' perceptions and practices regarding the use of vitamins and dietary supplements in the management of DN in different specialties.

Conclusion
This study showed that the constructed questionnaire had good levels of validity (content validity) and reliability that is able to cover different aspects of the current state of perception and practice among physicians regarding the management of DN. The preliminary results of the pilot study showed a good knowledge and perception of the respondents with DN. Even in the absence of a local guideline, the respondents followed the international guidelines in choosing anti-neuropathic agents however, they tended to add vitamin B complex/B12 as an adjuvant in their management plan.