|Year : 2021 | Volume
| Issue : 5 | Page : 485-492
Time-series forecasting analysis on the major treatment need among patients referred for periodontal and conservative treatments in IIUM Dental Outpatient Clinic
Azlini Ismail1, Zurainie Abllah2, Nur Aishah Muhammad Radhi3, Syazalina Musa4, Mohd Firdaus Akbar Abdul Halim5
1 Department of Fundamental Dental and Medical Sciences, Kulliyyah of Dentistry, International Islamic University Malaysia, Kuantan, Pahang, Malaysia
2 Department of Paediatric Dentistry and Dental Public Health, Kulliyyah of Dentistry, International Islamic University Malaysia, Kuantan, Pahang, Malaysia
3 Klinik Pergigian Slim River, Poliklinik Komuniti Slim River, Slim River, Perak, Malaysia
4 Klinik Pergigian Kuala Dungun, Dungun, Terengganu, Malaysia
5 Fakulti Sains Komputer dan Matematik, Universiti Teknologi MARA Cawangan Kelantan, Kota Bharu, Malaysia
|Date of Submission||06-Apr-2021|
|Date of Decision||06-Jun-2021|
|Date of Acceptance||23-Jun-2021|
|Date of Web Publication||11-Oct-2021|
Dr. Azlini Ismail
Department of Fundamental Dental and Medical Sciences, Kulliyyah of Dentistry, International Islamic University Malaysia, Indera Mahkota, 25200 Kuantan, Pahang.
Source of Support: None, Conflict of Interest: None
Aim: To identify the trend for major treatment needs among patients referred for conservative and periodontal treatments upon screening at Dental Outpatient Clinic, International Islamic University Malaysia, and to forecast the future need for this treatment using time-series forecasting analysis. Materials and Methods: This retrospective cross-sectional study used records of all patients referred for periodontal and conservative treatments from January 1, 2014 until December 31, 2016. The retrieved information includes patient’s age, gender, residential areas, and their treatment needs. Data were analyzed using the Statistical Package for the Social Sciences (SPSS) software program, version 20.0 (SPSS Inc., Chicago, IL, USA). The number of patients requiring each treatment need was counted, and the major treatment need was identified. For the time-series analysis, the weekly data for the major treatment need were fitted using few univariate models (moving-average, single-exponential, double-exponential, and Holt’s) in Minitab software version 17.0 (Minitab, LLC, State College, PA, USA) and the best-fitted model was chosen for forecasting the future need of this treatment. Results: From 3388 patients, majority were women (59.2%), aged 20–34 years old (54.5%), and came from urban areas (42.4%). The major treatment needs were scaling-polishing (65.0%) and dental filling (74.1%), respectively. The 3-year weekly pattern for both time series showed no apparent seasonal component. Using the best-fit moving-average model, it is forecasted that the 741 patients per year will require scaling-polishing and 855 patients per year will require dental-filling for the following years. Conclusions: The trends for scaling-polishing and dental-filling need among this population were nonseasonal, and the future needs were forecasted to remain high as in previous years.
Keywords: Conservative, Forecasting, Periodontal, Treatment Need, Trend
|How to cite this article:|
Ismail A, Abllah Z, Muhammad Radhi NA, Musa S, Abdul Halim MF. Time-series forecasting analysis on the major treatment need among patients referred for periodontal and conservative treatments in IIUM Dental Outpatient Clinic. J Int Oral Health 2021;13:485-92
|How to cite this URL:|
Ismail A, Abllah Z, Muhammad Radhi NA, Musa S, Abdul Halim MF. Time-series forecasting analysis on the major treatment need among patients referred for periodontal and conservative treatments in IIUM Dental Outpatient Clinic. J Int Oral Health [serial online] 2021 [cited 2021 Dec 6];13:485-92. Available from: https://www.jioh.org/text.asp?2021/13/5/485/327873
| Introduction|| |
Time-series forecasting was widely used in the health-care industry to project disease occurrences,,, to predict hospitalization rates and future visits to healthcare provider.,, Time-series forecasting analysis using data from 1977 until 2012/2013 has predicted a significant rise in dental visits from 2017 until 2040 in the United States and has helped the relevant authority in projecting their future dental care services in advance.
In the Malaysian context, approximately 94.0% of the Malaysian dentate adult population had periodontal diseases, whereas 84.6% of the Malaysian population had dental caries according to the National Oral Health Survey (NOHSA) 2010. In line with the national finding, Ismail et al. reported a considerable demand for conservative and periodontal treatments in a community study conducted among Kuantan residents in 2016. With the high cost of periodontal treatment, especially for periodontitis management as reported by Mohd Dom et al., these have become an economic burden for Malaysia’s oral health sector. Despite so, time-series forecasting analysis has not been widely used yet in Malaysia. In order to improve the provisions of periodontal and conservative treatment needs in this community, this study aimed to analyze the trend for major treatment need among patients referred for periodontal and conservative treatments during screening at the International Islamic University Malaysia (IIUM) Dental Outpatient Clinic and then to forecast the future need of this treatment in this study population using time-series forecasting analysis.
| Materials and Methods|| |
This study was conducted at the IIUM Dental Outpatient Clinic, a university-based dental institution providing dental services to the Kuantan community. Kuantan is the capital city of Pahang, a state located in East Peninsular Malaysia.
Study design and study population
This study involves three parts: the first part identified major treatment need among this study population, the second part was the time-series analysis, and the final part was the forecasting analysis. The study design for the first and the second parts of this study was a retrospective cross-sectional study, which involves extracting the secondary data from all patients referred for periodontal and conservative treatments from January 1, 2014 until December 31, 2016 at IIUM Dental Outpatient Clinic. It is compulsory to include all patients in this study as the time-series analysis requires information on the total number of patients across a timeline so that the data can be further used to forecast the number of patients with the same treatment need in future. As all patients need to be included, no sample size calculation was required for a time-series forecasting study. The retrieved information includes patient’s age, gender, residential areas, and their treatment needs. Treatment need in this study refers to the patients’ normative need as determined by the dental officers at this clinic after carrying out the routine screening and dental charting procedure for the patient’s oral condition. There was no risk of bias in this study as the data collection in this study only involves extracting data from the patient’s folder without direct or indirect interaction with the patients. There was no dropout in this study as all data used in this study were collected by the researcher from the Microsoft Excel (Microsoft, Redmond, WA, USA) databases available at the IIUM Clinic Outpatient. The database contains all patient’s information including the age, gender, residential areas, and the respective dental treatment need. The information was then cross-checked with the patient’s personal folders in hardcopy form, which were available in the clinic if there was any missing information in the database.
Inclusion and exclusion criteria
The inclusion criteria of the study were patients who attended IIUM Dental Outpatient Clinic and then referred for periodontal and conservative treatments from January 1, 2014 until December 31, 2016. The only exclusion criteria of the study were patients with missing information in their patient’s folder.
Identification of major treatment need among patients referred for periodontal and conservative treatments
To determine the major dental treatment need, patients who attended the clinic and were referred for periodontal and conservative treatment from January 1, 2015 until December 31, 2016 were included. The patients were classified based on their age group, gender, and residential area in Kuantan. The age was classified into five groups of 16 years old and below: 17–19 years old, 20–34 years old, 35–49 years old, 50–64 years old, and 65 years old and above. The patient’s residential areas were classified into five categories: urban areas in Kuantan, nonurban Areas in Kuantan, nonspecified areas in Kuantan, other districts in Pahang, and outside Pahang. The number of patients requiring each treatment needs was counted, and then the highest and the lowest treatment needs were identified. A descriptive analysis of the demographic profiles of patients was run in the Statistical Package for the Social Sciences (SPSS) software program, version 20.0.
Time-series analysis on the need for scaling-polishing and dental filling from 2014 to 2016 and forecasting the need for 2017 and 2018
For the time-series forecasting analysis, we have included patients who require scaling-polishing and dental filling from January 1, 2014 until December 31, 2016. The number of patients was counted weekly for a total number of 156 weeks. A simple weekly time plot was constructed for the number of patients requiring scaling-polishing and dental filling, and then a linear trend line was fitted. The time-series basic pattern and any unusual observation or characteristic was noted. The data were then fitted using time-series univariate models: “moving average,” “single exponential,” “double exponential,” and “Holt’s Method” models in MiniTab software version 17.0. The software then calculated the mean-squared error (MSE) and the mean absolute percentage error (MAPE) for each plotted model for two parts: the fitted part and the hold-out part. The univariate model with the least calculated MSE and MAPE was chosen to obtain the forecasted number of patients weekly for 2 years (2017 and 2018).
| Results|| |
Demographic profiles of patients who underwent screening at the IIUM Dental Outpatient Clinic and were referred for periodontal and conservative treatments (n = 3388) are tabulated in [Table 1]. The major age group of patients was from the age of 20–34 years old (54.5%), whereas the least number of patients was from the age group of 65 and above (1.4%). Female patients (59.2%) outnumbered male patients (40.8%). The majority of patients reside in the urban areas in Kuantan, Pahang (42.4%), followed by the nonurban areas within the Kuantan district (34.0%), and some did not specify the exact areas within the Kuantan district (4.5%). Some patients are residents of other districts in Pahang state, such as Bera, Chini, and Rompin (5.9%). Interestingly, the clinic also received patients from other states such as Terengganu and Selangor (13.2%).
|Table 1: Demographic profiles of patients referred for periodontal and conservative treatments at IIUM Dental Outpatient Clinic in 2015 and 2016|
Click here to view
Major treatment need among patients referred for periodontal and conservative treatments
Based on the 2-year records of patients referred for periodontal and conservative treatments in 2015 and 2016 [Figure 1], there were several types of treatment needs identified, and these include scaling-polishing, root planning, root canal treatment, impression, filling, minor oral surgery, extraction, oral health instructions, fixed prosthetics, prosthetics, pericoronitis, tobacco cessation, and oral medicine. The highest treatment need among patients referred for periodontal treatment was scaling-polishing (65.0%), whereas the lowest treatment need was the treatment of pericoronitis (0.1%) and tobacco cessation (0.1%) [Figure 1A]. For patients referred for conservative treatment, the highest treatment need was dental filling (74.1%), whereas the lowest treatment need was the treatment with oral medicine (0.1%) [Figure 1B]. It is important to note that the number of patients that requires extraction was relatively low compared to scaling-polishing and dental filling.
|Figure 1: Dental treatment need among patients at the IIUM Dental Outpatient Clinic referred for (A) periodontal and (B) conservative treatments in 2015 until 2016. |
Click here to view
Time-series analysis for the need of scaling-polishing and dental filling from 2014 to 2016 and forecasted values for 2017 and 2018
A simple weekly time plot on the number of first-visit patients that require scaling-polishing and dental fillings within 3 years from January 1, 2014 to December 31, 2016 is shown in [Figure 2]. A cursory observation indicates that the time series of these data was not stationary. Some outliers were observed wherein no patients referred for periodontal treatment required scaling-polishing at Week 2 until Week 5, Week 27, Week 31, Week 81, Week 85, Week 119, and Week 132. One unusually high plot was detected in Week 37 when 50 patients referred for periodontal treatment required scaling-polishing [Figure 2A]. In addition, there were also some outliers in which no patients referred for conservative treatment required dental fillings at Weeks 1 and 132, whereas in Weeks 23 and 146, there were an unusually high number of patients, 45 and 40 patients, respectively, that required dental fillings [Figure 2B]. The trend line for the need of scaling-polishing was minutely increased upward with a gradient of 0.0043 [Figure 2A], but the trend line was minutely declined downward for the need of filling with a gradient of –0.00328 [Figure 2B].
|Figure 2: Weekly attendance of patients at the IIUM Dental Outpatient Clinic referred for (A) scaling-polishing and (B) filling in 2014 until 2016. The dotted linear line represents the trend line|
Click here to view
[Table 2] and [Table 3] show the calculated MSE and MAPE from four different univariate models fitted with the 3-year data for scaling-polishing and filling needs. For the need for scaling-polishing [Table 2], the “moving average” model showed the lowest MAPE of 0.58 for the fitted part and the lowest MSE of 40.27, and MAPE of 1.35 for the hold-out part. The lowest MSE with the value of 26.69 for the fitted part was given by data plotted using Holt’s method. For the need for dental filling [Table 3], the “moving average” model showed the lowest MSE of 28.16 and MAPE of 0.37 for the fitted part and the MSE of 101.41 for the hold-out part. The lowest MAPE with the value of 1.49 for the hold-out part was given by data plotted using Holt’s method.
|Table 2: MSE and MAPE values using various models to forecast the need of scaling-polishing among patients referred for periodontal treatment|
Click here to view
|Table 3: MSE and MAPE values using various models to forecast the need of filling among patients referred for conservative treatment|
Click here to view
The forecasting was made using the “moving average” model as this model gives the lowest MSE and MAPE values for this data. Weekly numbers of patients required scaling-polishing and filling were fitted using the “moving average” model as shown in [Figure 3], and based on this pattern, the 2-year projections were made. Based on this “moving-average” model, it is forecasted that there will be 13 first-visit patients that were referred for periodontal treatment will require scaling-polishing per week; this is equivalent to 52 patients per month or 741 patients per year. Although for filling, this study forecasted that 15 first-visit patients referred for conservative treatment will require dental fillings per week; this equals to 60 patients per month or 855 patients per year.
|Figure 3: “Moving average” data plot and 2 years forecast for the need of (A) scaling-polishing and (B) filling. |
Click here to view
| Discussion|| |
In this study, most patients who underwent screening at IIUM Dental Outpatient Clinic and were referred for periodontal and conservative treatments were young adults, women, and those residing in urban areas of Kuantan. Young adults usually required conservative treatment in which most of them require dental filling to treat the cavity. As the age increases, the treatment need changed and was seemingly inclined towards prosthodontics, either for partial or complete dentures. Women outnumbered men in this study, and this was in line with the finding from the National Oral Health Survey Among Adults (NOHSA) 2010, in which a higher proportion of Malaysian adult women sought dental treatment as compared to men. The report also highlighted that women had significantly greater caries experience than men, and this trend was consistent in all age groups up to 74 years. Another university-based dental center in Malaysia similarly reported the strong tendency for women to visit their center. Among the plausible reasons are the changes in hormone levels in women; for instance, the increased levels of estrogen and a reduced saliva flow rate in women lessen the clearance of foreign material deposited on teeth and increases the dental caries rate. Besides, the high level of hormones in pregnant women’s blood and saliva may cause gingival reactions that may increase or cause gingival and periodontal disorders. Besides these physiological aspects of women, women are usually more attentive towards dental aesthetics than men, and for this reason, their dental visits are more frequent compared to men. Other than the age and gender factors, the clinic’s location in the urban area has provided better access for those residing in the urban areas. On the contrary, those residing in rural areas lack access and are thus less likely to visit dental service care providers.
Scaling-polishing was the major dental treatment need for patients referred for periodontal treatments in this study, possibly due to the high prevalence of calculus among this study population. This finding is consistent with the NOHSA (2000) and NOHSA (2010) reports, regarding the high prevalence of calculus with 56.9% in 2000 and 42.2% in 2010. Regular removal of plaque, calculus, debris, and staining deposits achieved through routine scaling-polishing treatment is considered prophylaxis for reducing gingivitis development and preventing periodontitis progression.
On the contrary, patients referred for conservative treatments in this study mainly required dental filling, possibly contributed by the high prevalence of dental caries among this study population. NOHSA (2000) reported a high prevalence of dental caries (90.7%) among adult populations in Malaysia. Dental caries or tooth decay may further progress into cavities if the necessary corrective or preventive action is not taken. Caries is usually treated by surgical removal of all carious tissue from the tooth and the placement of a restoration (filling) in the resulting cavity.
This study also showed that the number of first-visit patients referred for conservative and periodontal treatments and required extraction was relatively low compared to scaling-polishing and dental filling. This low number of patients requiring extraction was perhaps due to tooth decay degree, and the periodontal disease stages were not severe enough to cause extraction or minor oral surgery. In case of the need for extraction, the patients were usually directly referred for oral surgery. In addition, the current practice of minimal intervention dentistry also limits the removal of carious tooth tissue and emphasizes the repair and survival potential of the tooth tissue. Besides scaling-polishing, dental filling, and extraction, other dental treatment needs identified among the patients referred for periodontal and conservative treatments were root planning for preventive measures and oral medicines prescription. The usually prescribed oral medicines by dental practitioners include antibiotics and analgesics. The rest of the dental treatment needs were for prophylaxis, such as oral hygiene instruction and tobacco cessation. Oral hygiene instruction is essential as patients played a significant role in supra-gingival plaque control through meticulous oral hygiene, and this is actually integral for the success of any periodontal therapy. The noncompliance in oral hygiene efforts will usually result in unpredictable outcomes for surgical and nonsurgical periodontal treatment. In addition, tobacco cessation is also essential for patients undergoing periodontal treatment as the nonsurgical periodontal treatment outcomes usually improve after smoking cessation. The risk of getting periodontitis among smokers was 80% higher than quitters and nonsmokers.
After identifying the major dental treatment need among this study population, time-series forecasting was conducted to analyze the pattern of these major treatment needs in order to select an appropriate model for these data series and to use this as a model for time-series forecasting. This study showed that the number of patients requiring scaling-polishing and filling every week was always fluctuated around average. Despite several outliers were observed at certain weeks, the outliers were retained as removing the outlier may cause loss of vital information on the series being investigated. There was no seasonal component or cyclic behavior observed on the time-series as no regular fluctuations occurred quarterly or monthly within the 3-year period. However, the need for scaling-polishing was minutely increasing, and the need for filling was minutely declining in this 3-year period (2014 until 2016). The slightly-increasing and declining trend in this study might indicate the stagnant level of knowledge, awareness, and attitude towards oral health among this community.
The subsequent part of this study was the forecasting analysis, employing four univariate forecasting models, including moving-average, single exponential, double exponential and Holt’s method. These models were also being evaluated in other time-series forecasting analyses for projecting the upcoming demand of medicines, estimating the future demand of dentists and dental specialists as well as predicting future patient’s admission and visit, especially to the Emergency Department.,, Among these four univariate forecasting models, our time-series data were best fitted with the moving-average model as it gave the lowest MSE and MAPE values. MSE and MAPE were among the forecasting error parameters used to test for model accuracy. Low values of MSE and MAPE usually indicate good model performance and imply close matches between predicted and measured values.,, Hence, using this moving average model, both treatment needs in this study were forecasted to remain steadily high in the coming 2 years.
To the best of our knowledge, time-series forecasting analysis on the periodontal and conservative treatment need is by far still lacking. However, few time-series analyses on other dental treatment needs were present, but in different study populations. For instance, a study has shown a significant-declining trend in the use of amalgam restoration by dental students and professionals in an Australian university dental clinic within 10 years. Another study has conducted a time-series analysis on the number of patients seeking orthodontic treatment in one dental university hospital in South Korea from 2005 until 2015, and they have shown a distinct seasonal variation that peaked during summer and winter. Nonetheless, the scope of these studies was confined to the pattern and trend of time-series, without the forecasting component. Despite so, some time-series analyses include the forecasting component, but it was conducted for different purposes like projecting the future dental visit and predicting the forthcoming prevalence of certain oral conditions. For example, a study in the USA has predicted a significant rise in the total dental visits in that country, from 294 million in 2017 to 319 million in 2040, by analyzing the time-series data from 1977 until 2013. Meanwhile in China, time-series forecasting analysis has forecasted that the prevalence of early childhood caries in China will be decreasing in 2014 until 2017 by analyzing the previous prevalence data from 1988 until 2010.
However, projecting rates of dental care use or treatment need in the future is not a merely easy task because projections were made using available data, and forecasted trends may not persist if there is a major reform in the future. For instance, with the current coronavirus disease-2019 (COVID-19) pandemic that has affected the globe by the end of 2019, dental clinics are among the earliest offices closed by the higher authorities. This kind of change may significantly impact the forecasted values, thus having to be taken into consideration while conducting time-series forecasting. In addition to that, extending the time series is usually suggested to observe the trend’s consistency. Nevertheless, by looking at a broader perspective across a broader timeline, time-series forecasting analysis has assisted the respective authority in financial planning for provision of future dental services.
Time-series forecasting analysis can be extended to a much broader perspective covering a much more comprehensive dental treatment need at national level. This may assist the authorities in predicting future dental treatment need for an improved provision of oral healthcare services in Malaysia.
| Conclusion|| |
This study showed that the two major dental treatment needs among patients referred for periodontal and conservative treatments were scaling-polishing and dental filling, respectively. The trends for scaling-polishing and dental-filling among this population were nonseasonal, and the future needs for scaling-polishing and dental filling in this community were forecasted to remain high, as seen in the previous 3 years.
The authors would like to acknowledge Kuantan Municipal Council for providing the information on the residential areas in Kuantan, as well as the staff at the IIUM Dental Outpatient Clinic for assisting the team during data retrieving.
Financial support and sponsorship
This study was self-funded.
Conflicts of interest
There are no conflicts of interest.
AI was involved in designing research, coordinating research, results interpretation, and report writing. ZA was involved in results interpretation. NAMR and SM were involved in data collection and report writing. MFAAH was involved in data analysis.
Ethical policy and institutional review board statement
This study has obtained ethical approval from the International Islamic University Malaysia (IIUM) Research Ethics Committee with an approval number of IREC 762, on February 17, 2017. All the procedures have been performed as per the ethical guidelines laid down by Declaration of Helsinki (2013).
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consents that their information can be used for research purposes.
Data availability statement
The data set used in the current study is available upon request from Dr. Azlini Ismail (e-mail: [email protected]).
| References|| |
Ahmad WMAW, Noor NFM, Yudin ZBM, Aleng NA, Halim NA. Time series modeling and forecasting of dengue death occurrence in Malaysia using seasonal ARIMA techniques. Int J Public Health Clin Sci 2018;5:154-63.
Wang H, Tian CW, Wang WM, Luo XM. Time-series analysis of tuberculosis from 2005 to 2017 in china. Epidemiol Infect 2018;146:935-9.
Wagenaar BH, Augusto O, Beste J, Toomay SJ, Wickett E, Dunbar N, et al
. The 2014-2015 Ebola virus disease outbreak and primary healthcare delivery in Liberia: Time-series analyses for 2010-2016. PLOS Med 2018;15:e1002508.
Afilal M, Yalaoui F, Dugardin F, Amodeo L, Laplanche D, Blua P. Forecasting the emergency department patients flow. J Med Syst 2016;40:175.
Choudhury A, Urena E. Forecasting hourly emergency department arrival using time series analysis. Br J Health Care Manag 2020;26:34-43.
Gershon A, Thiruchelvam D, Moineddin R, Zhao XY, Hwee J, To T. Forecasting hospitalization and emergency department visit rates for chronic obstructive pulmonary disease. A time-series analysis. Ann Am Thorac Soc 2017;14:867-73.
Manski RJ, Meyerhoefer CD. Projecting the demand for dental care in 2040. J Dent Educ 2017;81:eS133-45.
Oral Health Division Ministry of Health. Oral Health Status Malaysian Adults2010. Putrajaya: National Oral Health Survey of Adults (NOHSA 2010).
Ismail A, Abllah Z, Radhi NAM, Musa S, Halim MFAA. Dental treatment needs among patients undergoing screening at a university-based dental institution in Kuantan, Pahang, Malaysia. Int J Orofac Health Sci 2020;1:18-27.
Mohd Dom TN, Ayob R, Abd Muttalib K, Aljunid SM. National economic burden associated with management of periodontitis in Malaysia. Int J Dent 2016;2016:1891074.
Jaafar A, Nasir WM, Ab Mumin N, Elias NNA, Sabri MAM. Reasons for seeking dental care among adults at an academic dental centre and the associated factors. Archiv Orofac Sci 2018;13:104-11.
Branch-Elliman D. A Gender-based Approach to Oral Health Changes Across the Lifespan. Philadelphia, PA: University of Pennsylvania; 2012. p. 1-59.
Srivastava A, Gupta KK, Srivastava S, Garg J. Effects of sex hormones on the gingiva in pregnancy: A review and report of two cases. J Adv Periodontol Implant Dent 2018;3:83-7.
Akbar FH, Pasinringi S, Awang AH. Relationship between health service access to dental conditions in urban and rural areas in Indonesia. Pesqui Bras Odontopediatria Clin Integr 2019;19:1-7.
Ogunbodede EO, Kida IA, Madjapa HS, Amedari M, Ehizele A, Mutave R, et al
. Oral health inequalities between rural and urban populations of the African and middle east region. Adv Dent Res 2015;27:18-25.
Oral Health Division Ministry of Health. Putrajaya: National Oral Health Survey of Adults (NOHSA 2000); 2000.
Worthington HV, Clarkson JE, Bryan G, Beirne PV. Routine scale and polish for periodontal health in adults. Cochrane Database Syst Rev2013;11:1-59.
Schwendicke F, Foster Page LA, Smith LA, Fontana M, Thomson WM, Baker SR. To fill or not to fill: A qualitative cross-country study on dentists’ decisions in managing non-cavitated proximal caries lesions. Implement Sci 2018;13:54.
Zhang Z, Zheng K, Li E, Li W, Li Q, Swain MV. Mechanical benefits of conservative restoration for dental fissure caries. J Mech Behav Biomed Mater 2016;53:11-20.
AbdulKader HK, Ali SM, Hassan MIA, Manan MM. Knowledge of prescribing antimicrobial among dental practitioners in Klang Valley region. Malays Dent J 2010;31:35-43.
Deas DE, Moritz AJ, Sagun RS Jr, Gruwell SF, Powell CA. Scaling and root planning vs. Conservative surgery in the treatment of chronic periodontitis. Periodontol 2000 2016;71:128-39.
Leite FRM, Nascimento GG, Baake S, Pedersen LD, Scheutz F, López R. Impact of smoking cessation on periodontitis: A systematic review and meta-analysis of prospective longitudinal observational and interventional studies. Nicotine Tob Res 2019;21:1600-8.
Alias M. Introductory Business Forecasting A Practical Approach. 3rd ed. Shah Alam: UPENA; 2011.
Bon A, Ng T. An optimisation of inventory demand forecasting in university healthcare centre. IOP Conf Ser: Mater Sci Eng 2017;166:1-11.
Tiwari R, Bhayat A, Chikte U. Forecasting for the need of dentists and specialists in South Africa until 2030. PLOS One 2021;16:e0251238.
Wargon M, Casalino E, Guidet B. From model to forecasting: A multicenter study in emergency departments. Acad Emerg Med 2010;17:970-8.
Calegari R, Fogliatto FS, Lucini FR, Neyeloff J, Kuchenbecker RS, Schaan BD. Forecasting daily volume and acuity of patients in the emergency department. Comput Math Methods Med 2016;2016:3863268.
Gul M, Celik E. An exhaustive review and analysis on applications of statistical forecasting in hospital emergency departments. Health Syst (Basingstoke) 2018;9:263-84.
Duwalage KI, Burkett E, White G, Wong A, Thompson MH. Forecasting daily counts of patient presentations in Australian emergency departments using statistical models with time-varying predictors. Emerg Med Australas 2020;32:618-25.
Lim HW, Park JH, Park HH, Lee SJ. Time series analysis of patients seeking orthodontic treatment at Seoul national university dental hospital over the past decade. Korean J Orthod 2017;47: 298-305.
Zhang X, Zhang L, Zhang Y, Liao Z, Song J. Predicting trend of early childhood caries in mainland China: A combined meta-analytic and mathematical modelling approach based on epidemiological surveys. Sci Rep 2017;7:1-9.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]