Knowledge of Primary Health Care Providers in Nairobi East District, Kenya, Regarding HIV-related Oral Facial and Other Common Oral Diseases and Conditions

Background: In the Kenya primary health care (PHC) setting where most patients, including nearly 1.4 million HIV-infected people, seek medical care, PHC providers are expected to identify and manage HIV-related oral diseases during general consultations. This study aimed to assess the Original Research Article current knowledge of clinical officers and nurses in Nairobi East district of Kenya regarding HIV-related oral diseases and conditions. Design and Methods: A 40-item questionnaire was used in interviewing all 57 PHC providers in 2 administrative divisions in the district in a cross-sectional survey. Assessed categories were: knowledge about HIV-related oral lesions, clinical appearance of HIV-suspected conditions, knowledge about oro-pharyngeal candidiasis (OPC), general dental knowledge, common appearances of OPC, knowledge about periodontitis, causes of dental caries, frequency of general oral examinations and past training in oral health topics. The first 4 categories were confirmed as sub domains, with Cronbach’s alpha of 0.57, 0.54, 0.59 and 0.45 respectively. Results: All 57 PHC providers (15 clinical officers and 42 nurses) completed the questionnaire (response rate 100%). PHC providers did not routinely perform oral examinations. Their knowledge about HIV-related oral health topics and general oral health was found to be generally inadequate. Recommendations: A training module on HIV-related oro-facial lesions for Nairobi PHC providers, incorporating a practical session covering oral examinations, is recommended; especially in this high HIV-prevalence environment.


INTRODUCTION
Nearly 90% of individuals infected with human immunodeficiency virus (HIV) develop visible signs of oral diseases in the course of their illness [1].The more common HIV-related oral lesions include oro-pharyngeal candidiasis (OPC), oral hairy leukoplakia, oral Kaposi's sarcoma, necrotising gingivitis and enlargement of the parotid glands [2][3][4][5][6][7].Some lesions appear soon after HIV infection, allowing early identification of this infection [3].The presence of these lesions may also serve as an early warning sign of non-response to treatment in patients already receiving highly active antiretroviral therapy (HAART), and of lowered immunological functions [8][9][10][11].Early identification and management of the lesions may eventually improve quality of care for these patients, as well as their quality of life.
In Kenya most patients, including an estimated 1.4 million HIV-infected ones [12], seek medical care in primary healthcare (PHC) settings [13,14].Clinical officers and nurses, the main service providers in PHC establishments, are trained to perform general medical examinations, make diagnoses and treat or refer patients to higher levels of care.As first-line service providers they are also expected to perform oral health tasks during consultations and record the results daily on government tally sheets [15].They need competences and skills to perform the expected oral healthcare tasks.However, these are assumed to be inadequate.To investigate their adequacy, the present study aimed to identify gaps in the knowledge of clinical officers and nurses regarding HIV-related oral lesions and other common oral conditions.It was conducted in the Nairobi East district.

DESIGN AND METHODS
Approval for this study was obtained from the Kenyatta National Hospital/University of Nairobi Ethics and Research Committee (approval number KNH-ERC/A/474) and from the Ministry of Public Health and Sanitation (Ref.NO.MPHS/IB/1/14 Vol.III).Written consent for the study to be conducted in Nairobi East district was received from the provincial and district heads of the department.This trial was registered in the Netherlands Trial Register (http://www.trialregister.nl,NTR2627).

Selection of Participants
Nairobi province as a capital city, has the largest number of clinical staff (medical doctors, dentists, nurses and clinical officers) in the country, with highly accessible public HFs, in comparison to other regions in the country [16,17,18,19].The proportions of these clinical staff nationally and in Nairobi East district are comparable [20], as shown in (Table 1).Of the total 466 clinical officers and nurses, 270 work in 54 public health facilities.The study was conducted among all 57 PHC providers in 4 health facilities (HFs) in Njiru (n=32) and 4 HFs in Makadara (n=25) divisions in the Nairobi East district.These were test and control divisions, as described in our earlier publication [21].dental researchers evaluated the appropriateness of the questionnaire.Questions were checked for simplicity and clarity and modified in accordance with their comments.The final questionnaire contained forty items (Fig. 1).The domain with questions pertaining to PHC provider knowledge concerning HIV-related orofacial lesions comprised seventeen items.The domain covering questions on knowledge about general oral conditions comprised fifteen openended items.Three items at the beginning of the questionnaire were related to background characteristics (age, gender and qualification).The last five items were contained in questions related to training, clinical experience and performance regarding oral examinations.This questionnaire is also discussed in our earlier publication [21].In generating an initial codebook the panel held a discussion in order to reach consensus on all possible correct and incorrect responses to open-ended questions with multiple responses.(Figs. 2 -22) show examples of several oral facial and other common oral diseases and conditions.

Implementation
The questionnaire was presented on the same day to all (n=57) PHC providers.All participants were given a brief spoken introduction to the confidentially coded questionnaire before they answered it.Participants were allowed to seek clarification regarding the questions.Each filled questionnaire was checked for completion before it was collected.Open-ended questionnaires with no right or wrong answer, such as those that needed 'further explanation' were noted, as were 'I do not know" responses.After coding of all scripts, results were checked for agreements and differences.Disagreements were discussed until consensus was reached.

Statistical Analysis
SAS software (SAS institute, Cary.NC, USA) was used for statistical analysis.The two aforementioned domains (knowledge about HIVrelated oro-facial lesions and knowledge about general oral health) contained 7 categories of questions.For each category mean values and standard deviations were calculated.Cronbach's alpha coefficients were used for checking internal consistency.

Characteristics of Participants
All 57 PHC providers (15 clinical officers and 42 nurses; response rate 100%) completed the questionnaire.There were more female nurses (n=32) and female COs (n=15) than male ones.Sixty percent of the participants were 40 to 52 years old and 40% were 23 to 39 years old.
Cronbach's alpha identified four sub-domains (from the 7 categories of questions), i.e., knowledge of symptoms of HIV-related oral lesions; knowledge of clinical appearance of HIVsuspected conditions; knowledge of OPC; knowledge about common appearances of OPC, and general dental knowledge (Table 2).

Knowledge of/about Symptoms of HIV-related Oro-facial Lesions
The mean domain score for PHC providers, relating to knowledge of common HIV-related oro-facial lesions; angular cheilitis, OPC, necrotising gingivitis/periodontitis, herpes zoster and Kaposi's sarcoma, was 4.6 (from maximum 7).This was higher than scores for the other three domains.The most highly scored item was related to the knowledge of herpes zoster as an HIV-related oro-facial lesion (0.94 out of maximum 1).

Knowledge of Clinical Appearance of HIV-suspected Conditions
Knowledge of three lesions -herpes simplex, herpes zoster and angular cheilitis was assessed.The mean score, 1.74 out of maximum possible 3, was considered moderate.Again, PHC providers were more familiar with herpes zoster than with the other two lesions.

Knowledge of OPC
Knowledge of signs and symptoms of pseudomebraneous OPC and reasons why this lesion was of significance in HAART patients was assessed.The moderate domain mean score of 1.84 out of maximum 3 was mainly derived from the PHC knowledge about clinical signs of pseudomebraneous OPC and associated symptoms such as pain experienced in eating and swallowing (0.78 and 0.68 respectively, out of maximum 1).The score on the clinical significance of this lesion as a marker of HAART failure was low (0.4 out of maximum 1).

Knowledge about Common Appearances of OPC, and General Dental Knowledge
Responses regarding three visual clinical aspects of OPC resulted in a mean score of 1.05 out of maximum 3. Knowledge about the pseudomebraneous type mainly contributed to the score.Eighteen percent knew 2 types: the second most-mentioned type of OPC being the erythematous type, while 12% of the respondents knew nothing about the clinical appearance of OPC.None of the PHC providers knew all three common clinical features.
Items in this domain related to questions commonly asked by community members regarding their oral health; such as those covering xerostomia, dental fluorosis and oral ulcerations and unusual patches in the mouth.The mean domain score of the PHCs was 2.66 out of maximum 6. Responses to causes of dental caries and signs and symptoms of periodontitis showed the lowest scores in the entire questionnaire (1.18 and 1.42, out of maximum 3, respectively).Most (79%), of the PHC providers knew of one or two factors that cause dental caries: mainly sugary foods and poor oral hygiene.Nearly 30% of the PHC providers did not know any of the three signs and/or symptoms of periodontal disease.

Frequency of Doing an Oral Examination
It was found that 41% of the PHC providers would perform an oral examination for a maximum of 10 out of any 50 patients who presented with an oral problem.Another 37% would examine between 10 and 40 of 50 patients.A minority, 22%, would examine more than 40 of 50 patients.

DISCUSSION
This study examined the knowledge of PHC providers, the main healthcare providers in the Nairobi PHC facilities, regarding common signs of HIV-related oro-facial lesions and other general oral conditions.

Knowledge of HIV-related Oro-facial Lesions
The results indicated that PHC providers had only moderate knowledge about HIV-related orofacial lesions.This was unexpected, as the current prevalence of HIV infection is high in Kenya [13].Moreover, the in-service training packages on management of HIV patients [23-25], routinely delivered to PHC providers, should have enhanced their skills.An expectation was that PHC providers would also know that OPC is not always a sign of HIV infection, since they handle patients with OPC in various clinics; such as child welfare and diabetes clinics.Contrary to this expectation, a large number of PHC providers (39%) still thought that OPC was always a sign of HIV infection, probably because this lesion is the one most frequently detected in HIV patients.
OPC is of great clinical significance, both in patients of unknown HIV status and in those already receiving HAART treatment.This lesion clinically presents as pseudomebraneous or erythematous types or as angular cheilitis, usually resulting in complaints of pain and difficulty in eating and swallowing [26].However, only a minority (22%) of the PHC providers routinely examined mouths of their patients.This percentage is lower than that of 32.2% observed among nurses in South Africa [27].Even in the intensive care unit setting, where oral care is part of the protocol for critical patient care, 21% of the nurses did not assess mouths of patients (28).
In similarity to other studies among nurses and dental students [27,29,30,31], PHC providers were more knowledgeable about OPC than about other (less frequent) HIV-related oral lesions.In this study it was noted that PHC providers were more familiar with the pseudomebraneous type than the other two types of OPC.Although the question read 'oropharyngeal candidiasis', the tongue was the most common site mentioned.It is possible that the PHC providers frequently examine the tongue when they routinely check for pallor.This agrees with the response of PHC providers: that in their routine practice they mainly examine teeth and swollen gums and sometimes, the tongue.It is evident that on the few occasions when the PHC providers examine the mouth, they miss out on the diagnosis of OPC when it is present in other parts of the mouth, such as the pharynx and the buccal mucosa, and when it is an erythematous type or angular cheilitis.Patients with HAART drug resistance often develop any type of OPC.This lesion may be common in this population, where inadequate patient access to HAART and nutritional support, as well tuberculosis (TB)/HIV co-infection, contribute to non-adherence and HAART drug resistance [32,33].Although this lesion is often managed in the HIV comprehensive care clinics as an opportunistic infection, PHC providers lacked knowledge about its clinical significance as a marker of HAART treatment failure.They mainly attributed its presence to side-effects of HAART.PHC providers in this setting miss opportunities to recognize HAART failure among their patients who develop OPC.
PHC providers did not suspect HIV infection in a child with enlarged parotid glands (diffuse bilateral non-painful swelling), even when the clinical history was suggestive of HIV exposure.Most associated it either with mumps or with a dental abscess.This gap between knowledge of HIV-related lesions and knowledge of the association of the lesions with HIV infection was also observed among dental students and oral hygienists [34].However, knowledge about the lesions and about association of the lesions with HIV infection was noted as being generally higher among dental students than among nurses [29,[34][35][36].Given the high prevalence of HIV infection in this community, HIV-related lesions that were assessed were likely to have been identified in the consulting rooms of the PC providers.Agbelusi and Wright, 2005 [37] showed that laboratory tests confirmed that nearly all (92%) patients identified in a dental clinic as having HIV-related oral lesions were HIV-infected.Moderate, and in some aspects low, knowledge of PHC providers in identifying these lesions, and lack of association of the lesions with HIV infection, indicates missed opportunities for HIV testing and early detection and treatment in this setting.Moreover, nearly all PHC providers said that they never performed an extra-oral examination in routine practice.

Knowledge of General Oral Health
Assessment in this domain was related to questions that patients and community members commonly ask health personnel regarding their oral health.These included dental caries; periodontal diseases, mouth ulcerations and dental fluorosis, which PHC providers are expected to record in the outpatient tally sheet.In similarity to others [27,36,38,39], this study indicated that PHC providers in this setting lacked sufficient knowledge on oral health topics.Their knowledge regarding general oral health was much lower than for HIV-related oral lesions.However, it was noted that the majority, 80%, knew that sugar and poor oral hygiene contribute to dental caries.Information on the importance of good oral health could be integrated into daily health talks in PHC facilities, especially in antenatal care and child welfare clinics, to prevent dental decay in children.
Although all patients who seek general consultations with the PHC providers should ideally be orally examined, the choice of patients was narrowed down to those presenting complaints that warranted an oral examination, e.g., pain experience when chewing and swallowing, skin rashes and febrile conditions.
The low scores suggested that extra-oral examinations were rarely performed.PHC providers reported that when they occasionally did intra-oral examinations, they generally looked at the condition of teeth and gums and were therefore likely to miss other oral pathologies.In personal communication with LNK, the PHC stated that they examined the throat only when the patient had a suggestive history, such as 'upper respiratory tract infection' and tonsillitis.In an earlier study [40], PHC providers indicated that they were willing to do oral examinations but failed to do so, owing to lack of basic tools such as spatulas, mouth masks and torches.Results of the present study show that they also lacked knowledge about performing expected oral healthcare tasks.
The PHC providers reported that oral health topics were scantily covered in their pre-service training (some of the short in-service courses that they regularly receive, such as the integrated management of adulthood illnesses [25], include an oral health module on HIV care).Apparently, the training modules do not adequately cover the practical skills aspects that the PHC providers need for identifying and managing HIV-related oral lesions.This is partially understandable, as recognition of dental diseases needs much experience and training.PHC providers cannot be expected to gain sufficient knowledge and insight regarding dental diseases in short training courses.A training module for the PHC providers, incorporating a practical session on how to perform a basic oral examination and recognise (HIV-related) oral lesions is recommended, to build their skills and competences.

Significance for Public Health
Immunological impairment in HIV patients often predisposes them to HIV-related oral lesions.Therefore, comprehensive care of these patients needs healthcare workers to include oral healthcare of the patients, at all levels of care.Targeted health personnel at PHC level are in many cases non-dental professionals who are expected to integrate diagnosis and management of oral lesions into their routine practice.These health workers need much training and sufficient skills to perform the expected tasks.
Integrating oral health care into PHC is a global and a national oral policy objective, which calls for development of a 'special curriculum for existing training programs for special groups'.As no questionnaire was available in the literature, assessment of knowledge gaps was the first important step in tailor-making a training program for PHC providers.The questionnaire that was used can be adapted in different settings.

CONCLUSION
Knowledge of primary health care providers in Nairobi East District, Kenya, regarding HIVrelated oral facial and other common oral diseases and conditions was found to be generally inadequate.Peer-review history: The peer review history for this paper can be accessed here: http://www.sciencedomain.org/review-history.php?iid=672&id=12&aid=6291

Table 1 . Comparison of proportions of clinical staff (medical doctors, dentists, nurses and clinical officers) Kenya: countrywide and in Nairobi east district in 2009
Source: * [16;17], **[20]All clinical officers and nurses offering clinical services were included.A 4-year diploma course in clinical medicine provides basic training for clinical officers.Registered and enrolled nurses complete 3.5 and 2 years of training at diploma and certificate levels, respectively.
Trained data clerks entered raw data into an Excel file.A random check on the entered data was done by LNK.Two experienced dentists (LNK and ED) coded all answers to open-ended questions.The codebook was improved through use of selected questionnaire responses[22].

Table 2 . Cronbach's alpha, mean, standard deviation, minimum and maximum scores by domains retrieved from the questionnaire on PHC providers' knowledge of HIV-related oral facial lesions and other common oral lesions and conditions Domain/ Categories of questions Cronbach's alpha Mean Std dev Minimum Maximum Knowledge on HIV-related oro-facial diseases
*sub domain