Pa ge 1 Pa ge 1 American Journal of Applied Statistics and Economics (AJASE) Effective Health Care Plan for National Health Insurance Scheme Patients with Non-Communicable Diseases in Plateau North Senatorial District Philemon Polycarp Davwar1* Volume 2 Issue 1, Year 2022 https://journals.e-palli.com/home/index.php/ajase Article Information ABSTRACT Received: February 25, 2023 Accepted: May 08, 2023 Published: May 14, 2023 Patients of non-communicable diseases (NCDs) are usually placed on a life-long prescription or procedure. This is sometimes a problem in itself. They could get tired since they are in most cases, not sickly, they could get careless as a result of familiarity or boredom, etc. The greatest challenge is that they could decide not to access care or continue to access care at a particular location for several reasons which may be culturally, politically, or socially influenced. Consequently, their situation could become more complicated. The location of access points to healthcare for them becomes a critical issue here. In this research, we looked at location of healthcare access points for NCD patients living in Plateau North Senatorial district who are also registered with the National Health Insurance Scheme (NHIS). Data on specialist opinions as to the relevance of each specialist and equipment for the effective management of the associated NCD was collected and analysed to determine relevance ratings. These ratings along with data on availability of specialists and equipment from each care provider registered with the NHIS in the study area, was analysed using the Patient- based Set Covering Location Model (Davwar, Wajiga & Okolo, 2021). The results show care providers that can provide healthcare services to such patients at a quality level T, the threshold value for patients in each location, below which a service provider is not considered. This was done for each NCD (Diabetes and Cancer) under consideration. In the analysis, each NCD was considered a scenario. For each patient of diabetes and or cancer, in each location in the study area, this research determines points of access to quality care. Keywords Patient-Based Facility Location, Non-Communicable Diseases, Diabetes, Cancer, Set Covering Location Models 1 Department of Mathematics and Statistics, Federal Polytechnic Idah, Kogi State, Nigeria * Corresponding author’s e-mail: ppdavwar@gmail.com INTRODUCTION Patients of chronic non-communicable diseases are usually placed on a life-long prescription or procedure. This is sometimes a problem in itself. They could get tired since they are in most cases, not sickly, they could get careless as a result of familiarity or boredom, etc. The greatest challenge is that they could decide not to continue to access care at a particular location for a number of reason which may be cultural, political, or social. They are mostly healthy looking and so may feel uncomfortable that other people know that they are sick. They may prefer to pretend they have no problems. They most times prefer to access medical care in some concealed manner. So this study attempts to locate points of access to healthcare service for them. Since this is a medical problem, it is required that a total coverage is achieved. Ordinarily, the classical set covering facility location model would suffice, except that if the classical model optimizes well, a patient will be assigned to one point of healthcare service only. Therefore, if this patient has other psychological considerations but compelled to access care in this one facility prescribed by the model, he may do so but suffer from a psychological stress which may complicate the management of his case over time. The objective of this research is to locate points of access to health care services for patients suffering from either diabetes or Cancer and who reside in any of the local government areas in Plateau North senatorial district. These points of access provide ‘cover’ for each of these patient locations at a quality level of at least a threshold value as prescribed by the patient-based facility location model (Davwar, Wajiga & Okolo, 2021). LITERATURE REVIEW NCDs results from a mixture of genetic, physiological, environmental and behavioural factors with pronounced dangers because of its chronic nature. Annually, its global mortality is 41 million people, which accounts for 70% of global deaths. Approximately 40% of these deaths occur among people aged between 30 and 69 years (WHO, 2018a), while 80% of these early deaths occur in low and middle-income countries (WHO, 2018b). Global Health Observatory data in 2018 predicted that deaths from NCDs would rise to about 52 million worldwide in the year 2030 (WHO, 2020). Of all NCDs, cardiovascular disease accounts for about 40% of all deaths annually while cancers, respiratory diseases and diabetes account for 22%, 10% and 4% respectively. These four diseases similarly account for over 80% of all premature deaths (Olukoya, 2017). In a WHO report, the probability of dying prematurely from NCDs in Nigeria is put at 20% (WHO, 2018c) while the projected prevalence estimate of diabetes in Nigeria is 4.04% (IDF, 2011). According to the 2012 Globocan data, Nigeria’s top five cancer burdens are breast, cervix uteri, liver, prostate and colorectal cancers (Awodele, Adeyomoye, Awodele, Fayankinnu & Dolapo, 2011). Workable and evidence-based solutions must be provided (Ezzati & Riboli, 2012) to relieve the burden of NCDs in Nigeria which is the aim of this research. Current, https://journals.e-palli.com/home/index.php/ajahs mailto:ppdavwar@gmail.com Pa ge 2 https://journals.e-palli.com/home/index.php/ajase Am. J. Appl. Stat. Econ.1(1) 1-6, 2023 Daskin, & David (2001) maintained among others that location decisions are often strategic in nature, frequently impose economic externalities and often extremely difficult to solve. He further puts it that there does not exist a general location model that is appropriate for all potential or existing applications. Therefore, different models are developed for different location decisions. The model employed in this study Davwar, Wajiga & Okolo (2021) was developed to cater for the peculiarities of patients with NCDs. It will be recalled that Diabetes and Cancer are both NCDs. They are probably the most common chronic health challenges in the contemporary society. Any research that could lead to an enhanced management of NCDs is very needful because NCDs are an important contemporary health issue, and is growing in importance, because: i. A person’s social circumstances affect the chance of him/her having a NCD greatly. So, the chances are right that more people will come down with one NCD or the other. ii. Some patients have multiple NCDs, which make their care particularly complex. iii. NCDs usually have a mild beginning, a simple social habit, or what appears to be a normal life, but gradually grow into a life-threatening monster. iv. There is evidence that NCDs can be better managed through increased ease of access to special medical help. METHODOLOGY Data from Plateau North Senatorial district on Diabetes and Cancer were collected for this study. Data on availability of specialists and equipment/procedures were collected from all NHIS service providers in the district and used for the analyses. The Microsoft EXCEL solver was used to analyse the data. The analyses and results are presented below: The Patient-Based Set Covering Facility Location model developed by Davwar, Wajiga & Okolo (2021) was used in this analyses. Given a set of NHIS service providers in the study area, Ni= {J\Qjis≥Ts }, ∀j ∈ J, ∀i ∈ I and for each scenario s, s= 1,2 And given that: Ni is a set of facilities capable of offering service to patients in location i with scenario s at a quality level of at least Ts Where: Ts= Quality threshold for scenario s. Qji= 1/dij{Fjs+ Ejs } (1) Fjs= Rating of Specialists at service point j for management of scenario s. Ejs= Rating of equipment at service point j for management of scenario s. Qjis= Quality of service facility j can provide to patient location i with scenario s. And dij= Distance from patient location i to hospital j Then the minimum number of health facilities Zs that can provide coverage for patients with chronic condition RESULTS AND DISCUSSION To evaluate equation (1) we require the following: The Specialists and Equipment/Procedure Ratings (Fjs and Ejs) respectively for every specialist and equipment/procedure Using the data collected a score was determined for each specialist and equipment/procedure respectively, as a measure of their relevance to the management of a particular condition (scenario). The rating which is from a five-point scale is the average rating from the responses of professionals in the field of health care provision (Doctors, Laboratory Scientists, Pharmacists, others) on the relevance of a given specialist or equipment/ procedure in the management of the particular chronic challenge as presented in table 1 below: Table 1: Showing Specialists and Equipment/Procedure Ratings Scenario Specialists Fjs Equipment Ejs 1. Diabetes Dietician 5.00 Spectrophometer 4.83 Endocrinologists 4.80 Glucometer 4.80 Pharmacists 4.80 Insulin Infusion Pump 4.08 Lab. Scientists 4.71 Emergency Rm Med 4.50 Dialysis Nurses 3.45 2. Cancer Oncologists 4.97 Radiation Machines 4.66 Pathologists 4.59 Microscopes 4.53 Surgeons 4.46 CT Scan MRI 4.24 s in all patient locations is given as: Where: Zs= minimum number of facilities required to offer complete coverage to all patient locations with scenario s. J= the set of eligible NHIS facilities (Indexed by j) I= the set of patients’ locations (Indexed by i) https://journals.e-palli.com/home/index.php/ajase Pa ge 3 https://journals.e-palli.com/home/index.php/ajase Am. J. Appl. Stat. Econ.1(1) 1-6, 2023 Pharmacists 4.44 Immunohistochemistry 4.21 Psychologists 3.47 X-Ray Machine 3.83 Psychotherapists 3.15 *Key to abbreviations in table above: CT Computerized Tomography, MRI Magnetic Resonance Imaging Determination of Health Service Centre Potentials (Total Scores for Specialists and Equipment/ Procedure Availability (Fjs+ Ejs)) Data on the availability of specialists and equipment/ procedure at each facility was collected and used to determine specialists’ availability score Fjs and equipment/procedure availability score Ejs (ie Potential) for each facility as presented in table 2 below: Table 2: Hospital Potentials (Total Scores for Specialists and Equipment/Procedure Availability (Fjs+ Ejs) S/N HOSPITAL DIABETES CANCER 1 PL/0057 14.29 14.02 2 PL/0163 14.29 14.02 3 PL/0003 14.29 14.02 4 PL/0172 14.29 14.02 5 PL/0171 14.29 14.02 6 PL/0150 27.21 30.78 7 PL/0014 14.29 14.02 8 PL/0103 14.29 17.86 9 PL/0159 14.29 14.02 10 PL/0116 14.29 14.02 11 PL/0169 14.29 14.02 12 PL/0100 41.13 40.60 13 PL/0168 14.29 14.02 14 PL/0010 14.29 17.86 15 PL/0067 14.29 14.02 16 PL/0079 14.29 14.02 17 PL/0021 14.29 14.02 18 PL/0008 14.29 14.02 19 PL/0075 22.71 26.28 20 PL/0104 14.29 17.86 21 PL/0170 14.29 14.02 22 PL/0009 14.29 17.86 23 PL/0076 14.29 14.02 24 PL/0064 14.29 14.02 25 PL/0120 14.29 14.02 26 PL/0176 14.29 14.02 27 PL/0058 14.29 14.02 28 PL/0019 14.29 14.02 29 PL/0018 14.29 14.02 30 PL/0115 14.29 14.02 31 PL/0111 14.29 14.02 32 PL/0077 14.29 14.02 33 PL/0020 14.29 14.02 34 PL/0066 14.29 17.86 35 PL/0158 28.21 31.94 36 PL/0007 14.29 14.02 37 PL/0057 14.29 14.02 https://journals.e-palli.com/home/index.php/ajase Pa ge 4 https://journals.e-palli.com/home/index.php/ajase Am. J. Appl. Stat. Econ.1(1) 1-6, 2023 38 PL/0153 14.29 14.02 39 PL/0023 14.29 14.02 40 PL/0108 14.29 14.02 41 PL/0117 14.29 14.02 42 PL/0080 14.29 14.02 43 PL/0139 14.29 14.02 44 PL/0148 14.29 14.02 45 PL/0161 14.29 14.02 46 PL/0165 14.29 14.02 47 PL/0106 14.29 17.86 48 PL/0122 14.29 14.02 49 PL/0164 14.29 14.02 50 PL/0107 14.29 14.02 51 PL/0167 14.29 14.02 52 PL/0102 23.79 27.36 53 PL/0162 14.29 14.02 54 PL/0141 14.29 17.86 55 PL/0121 17.71 21.28 56 PL/0001 14.29 14.02 57 PL/0071 14.29 14.02 Note: See Appendix III for key to Hospital Codes. Determination of the Quality of Facility j to Handle Patients In Location i With Scenario s. (Qjis) We required the distances from the patient location to the various potential facilities (Appendix II) and the potentials (∑(Ejs+Fjs)) of these facilities to handle each of the chronic conditions. The potentials were appropriately combined with the distance factors (dij) to form the quality level of facility j to handle patients in location i, with scenario s. This quality level of facility j to handle patients in location i, with scenario s is computed as distance weighted. This is because of the negative effect distance has on the quality of service. It is computed for the two scenarios (Diabetes and Cancer) therefore as: Qjis= 1/( dij ) {Ejs+ Fjs } And presented in 3A (for Diabetes) and 3B (for Cancer) below: Table 3: Quality level of Facility j to Handle Patients in Location i With Scenario s. (Qjis) Diabetes Cancer Hospital B as sa Jo s N th Jo s E st Jo s St h B /L ad i R iy om B as sa Jo s N th Jo s E st Jo s St h B /L ad i R iy om PL/0057 4.76 0.41 0.19 0.27 0.18 0.17 4.67 0.40 0.19 0.27 0.18 0.17 PL/0163 4.76 0.41 0.19 0.27 0.18 0.17 4.67 0.40 0.19 0.27 0.18 0.17 PL/0003 4.76 0.41 0.19 0.27 0.18 0.17 4.67 0.40 0.19 0.27 0.18 0.17 PL/0172 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0171 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0150 0.78 9.07 0.70 1.60 0.60 0.58 0.88 10.26 0.79 1.81 0.68 0.65 PL/0014 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0103 0.41 4.76 0.37 0.84 0.32 0.30 0.51 5.95 0.46 1.05 0.40 0.38 PL/0159 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0116 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0169 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0100 1.18 13.71 1.05 2.42 0.91 0.88 1.16 13.53 1.04 2.39 0.90 0.86 PL/0168 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0010 0.41 4.76 0.37 0.84 0.32 0.30 0.51 5.95 0.46 1.05 0.40 0.38 https://journals.e-palli.com/home/index.php/ajase Pa ge 5 https://journals.e-palli.com/home/index.php/ajase Am. J. Appl. Stat. Econ.1(1) 1-6, 2023 PL/0067 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0079 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0021 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0008 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0075 0.65 7.57 0.58 1.34 0.50 0.48 0.75 8.76 0.67 1.55 0.58 0.56 PL/0104 0.41 4.76 0.37 0.84 0.32 0.30 0.51 5.95 0.46 1.05 0.40 0.38 PL/0170 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0009 0.41 4.76 0.37 0.84 0.32 0.30 0.51 5.95 0.46 1.05 0.40 0.38 PL/0076 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0064 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0120 0.41 4.76 0.37 0.84 0.32 0.30 0.40 4.67 0.36 0.82 0.31 0.30 PL/0176 0.19 0.37 4.76 0.30 0.19 0.17 0.19 0.36 4.67 0.29 0.19 0.17 PL/0058 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0019 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0018 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0115 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0111 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0077 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0020 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0066 0.27 0.84 0.30 4.76 0.51 0.38 0.34 1.05 0.37 5.95 0.64 0.47 PL/0158 0.54 1.66 0.59 9.40 1.01 0.74 0.61 1.88 0.67 10.65 1.14 0.84 PL/0007 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0057 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0153 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0023 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0108 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0117 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0080 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0139 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0148 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0161 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0165 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0106 0.27 0.84 0.30 4.76 0.51 0.38 0.34 1.05 0.37 5.95 0.64 0.47 PL/0122 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0164 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0107 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0167 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0102 0.46 1.40 0.50 7.93 0.85 0.63 0.53 1.61 0.57 9.12 0.98 0.72 PL/0162 0.27 0.84 0.30 4.76 0.51 0.38 0.27 0.82 0.29 4.67 0.50 0.37 PL/0141 0.27 0.84 0.30 4.76 0.51 0.38 0.34 1.05 0.37 5.95 0.64 0.47 PL/0121 0.34 1.04 0.37 5.90 0.63 0.47 0.41 1.25 0.44 7.09 0.76 0.56 PL/0001 0.18 0.32 0.19 0.51 4.76 0.30 0.18 0.31 0.19 0.50 4.67 0.30 PL/0071 0.18 0.32 0.19 0.51 4.76 0.30 0.18 0.31 0.19 0.50 4.67 0.30 Determination of the Quality Threshold (T) The threshold value T is determined as the lowest quality level a service point can offer a demand point to qualify that service point for consideration as a candidate in the analysis for the scenario under consideration. It therefore varies from scenario to scenario and is set by experiment as the smallest quality value for which the model has a feasible solution. https://journals.e-palli.com/home/index.php/ajase Pa ge 6 https://journals.e-palli.com/home/index.php/ajase Am. J. Appl. Stat. Econ.1(1) 1-6, 2023 Results from Model Analyses The results from the model analyses shows the hospitals selected and the patient location that can enjoy service from them for each of the two scenarios (Diabetes and Cancer respectively) at a quality level of at least T (the quality threshold for that scenario). These results are here presented on the table below. The results shown below are identical. This is a pure Table 4: Showing results of the Model Analyses Diabetes Cancer Facility Patient Location Facility Patient Location Jo s N or th Jo s So ut h Jo s E as t B /L ad i B as sa R yo m Jo s N or th Jo s So ut h Jo s E as t B /L ad i B as sa R yo m PL/0150 ok ok ok ok ok ok PL/0150 ok ok ok ok ok ok PL/0100 ok ok ok ok ok ok PL/0100 ok ok ok ok ok ok PL/0075 ok ok ok ok ok ok PL/0075 ok ok ok ok ok ok PL/0158 ok ok ok ok ok ok PL/0158 ok ok ok ok ok ok PL/0102 ok ok ok ok ok ok PL/0102 ok ok ok ok ok ok *Key: OK Means Covered by the Corresponding Facility coincidence and does not mean that other scenarios will have identical results. From the above results, a patient in any of the six patient locations (i = 1,2,…6) can access service from any of the five hospitals (j = 1,2,…5) chosen by the model with a quality level of at least T (T= 0.2 for Diabetes and T=0.24 for Cancer). CONCLUSION This study therefore concludes that the five hospitals (Bingham University Teaching Hospital, Jos University Teaching Hospital, Our Lady of Apostles Hospital, Dee Medical Centre and Plateau Specialist Hospital) shown in the results provide adequate coverage (makes accessible) to patients with any of the chronic conditions (Diabetes or Cancer) who reside in any part of the Plateau North Senatorial District and who are NHIS subscribers. RECOMMENDATION The study recommends any of the five identified hospitals to NHIS subscribers (patients) having Diabetes or Cancer and who reside in any part of the Plateau north senatorial district. Policies that allow the establishment of community parks, sidewalks, bike lanes, playgrounds or village square areas with beautiful landscapes where people can gather, jog, meet and play during leisure encourages people to participate in physical activity. Acknowledgements I would like to acknowledgement the advice and encouragement received from Professor G. Wajiga and Professor. H. G. Muazu all of Modibbo Adama University Yola, Adamawa State, Nigeria. REFERENCES Awodele, O., Adeyomoye, A. A., Awodele, D. F., Fayankinnu, V. B., and Dolapo, D. C. (2011). Cancer distribution pattern in south-western Nigeria. Tanzan J Health Res., 13(2), 125-31. https://doi.org/10.4314/ thrb.v13i2.55226. Current, J., Daskin, M., and David, S. (2001). Discrete network location models, in Drezner, Z. and Hamacher, H. W., (eds). (2002). Facility location: Applications and theory, Berlin: Springer, 81-118. Davwar, P. P., Wajiga, G. M. and Okolo, A. (2021). A patient-based set covering facility location model, Global Scientific Journals, 9(10), 2423-2428 Ezzati, M., and Riboli, E. (2012). Can non-communicable diseases be prevented? Lessons from studies of populations and individuals. Science, 337(6101), 1482- 7. https://doi.org/10.1126/science.1227001 International Diabetes Federation (IDF, 2011). IDF Diabetes Atlas. 5th ed. Brussels: International Diabetes Federation. Olukoya, O. (2017). War against non-communicable disease: How ready is Nigeria? [eds], Annals of Ibadan Postgraduate Medicine, 15(1). World Health Organization (2018a). Non-communicable diseases. Retrieved on April 12, 2020. https:// www.who.int/news-room/fact-sheets/detail/ noncommunicablediseases. World Health Organization (2018b). Non-communicable diseases progress monitor 2018. Retrieved on April 12, 2020. http://apps.who.int/iris/ bitstream/10665/258940/1/9789241513029eng. pdf?ua51. World Health Organization (2018c). Global health observatory data 2018: Premature NCD deaths. Retrieved on April 12, 2020. https://www.who.int/ gho/ncd/mortality_morbidity/ncd_premature_text/ en/. World Health Organization (WHO, 2020). Non- communicable diseases. [cited 2020 April 12]. Available from: http://www.who.int/mediacentrefactsheets/ fs355/en/. https://journals.e-palli.com/home/index.php/ajase