Publication: Sağlık Sektöründe Veri Zarflama Analizi İle Etkinlikölçümü: Samsun İli Örnek Uygulama
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Abstract
Sağlık hizmetleri toplumsal yaşamı, özellikle de kişilerin yaşamını direkt olarak etkilemektedir. Sağlık sektörünün temelinde yer alan hastanelerin finansmanı kamu veya özel kaynaklardan sağlanmaktadır. Kamu hastanelerinin kaynakları ne kadar verimli kullandıkları sağlık politikalarının oluşturulmasında hayati öneme sahiptir. Veri Zarflama Analizi (VZA) özellikle hizmet sektöründe faaliyet gösteren aynı amaç ve hedeflere sahip işletmelerin göreceli olarak verimliliğini ölçmede yaygın olarak kullanılan doğrusal programlamanın özel bir uygulama şeklidir. Geleneksel VZA modelleri, yalnızca kullanılan girdi ve üretilen çıktıların kesin olarak bilindiği durumlarda uygulanabilmektedir. Verilerin kesin olarak bilinmediği durumlarda etkinlik ölçümlerinin yapılabilmesi için ise Bulanık Veri Zarflama Analizi (BVZA) modelleri geliştirilmiştir. Bu çalışmada, Samsun Kamu Hastaneleri Birliği'ne bağlı 18 tane kamu hastanesinin etkinlikleri ölçülmüştür. Hastanelerin etkinlik ölçümünde en uygun girdi ve çıktı veri setinin oluşturulmasında elde edilen veriler ve literatürden yararlanılmıştır. Etkinlik ölçümü için 4 adet girdi (Doktor Sayısı, Poliklinik Sayısı, Servislerdeki Toplam Yatak Sayısı, Toplam Yoğun Bakım Yatak Sayısı,) ve 4 adet çıktı (Toplam Muayene Sayısı, Toplam ameliyat sayısı, Toplam Yatılan Gün Sayısı, Toplam Yatan Hasta Sayısı) belirlenmiştir. Araştırmada kullanılan değişkenler ile ilgili veriler; Sağlık Bakanlığı, Kamu Hastaneler Birliği Genel Sekreterliği'nden resmi olarak temin edilmiş olup hastanelerin 2014- 2015- 2016 ve 2017 yıllarına ait her yıl için 12 aylık veriler tespit edilmiş, ancak değerlendirmede veriler yıllık dönemler halinde kullanılmıştır. Çalışma kapsamında değerlendirilen hastanelerin (Karar Verme Birimi-KVB) etkinlik skorlarının belirlenmesinde ölçeğe göre değişken (VRS) ve ölçeğe göre sabit getiri (CRS) varsayımı altında hem girdi yönelimli hem de çıktı yönelimli VZA modeli kullanılmıştır. Ayrıca, çalışma kapsamında bulanık veri zarflama (BVZA) modellerinden Saati- Memariani modeli kullanılmıştır. Samsun'da doktor başına düşen muayene sayısı, doktor başına düşen kişi sayısı, doktor başına düşen muayene sayılarının Türkiye ortalamsının üzerinde olduğu belirlenmiştir. Ayrıca çalıma kapsamında Veri Zarlama Analizi (VZA) için kullanılacak girdi ve çıktı değişkenleri ararasında potitif yönlü kuvvetli ilişkinin olduğu tespit edilmiştir. Çalışmada dikkate alınan karar verme birimleri-KVB'den (18 adet kamu hastanesi) hem girdi odaklı hemde çıktı odaklı CCR modeline göre büyük bir kısmının (12 hatane) etkin olduğu tespit edilmiş olup, 2014'den 2017 yılına doğru ise etkinlik değerlerinde kısmen artışların olduğu görülmüştür. BCC modelinde ise KVB'lerinin Girdi odaklı BCC modeline göre; 2014 yılında 16 hastane, 2015 yılında 15 hastane, 2016 yılında 14 hastane etkin, 2017 yılında ise 15 hastanenin etkin olduğu belirlenmiştir. Yine, çıktı odaklı BCC modeline göre ise; 2014 yılında 16 hastane etkin, 2015 yılında 15 hastane etkin, 2016 yılında 14 hastane etkin, 2017 yılında ise 15 hastanenin etkin olduğu görülmüştür. Çalışma kapsamında değerlendirilen karar verme birimlerinin etkinlik değişimlerininde yıllar içerisinde farklılık gösterdiği tespit edilmiştir. Karar verme birimlerinin etkinliklerinin belirlenmesinde kulllanılan bulanık veri zarflama (BVZA) analizlerinden Saati- Memariani modeli sonuçlarına göre ise her KVB'nin belirlenen alfa kesim kümlerine göre ortama, alt sınır ve üst sınır değerlerine göre etkinlik değerlerinin farklı olduğu belirlenmiştir. Özellikle iyimser yaklaşım olan üst sınır değerlerine göre tüm KVB lerinin etkin olduğu belirlenmiştir.
Health services are directly affecting social life, especially people's life. The funding of hospitals based at the healthcare sector is provided by public or private sources. Public hospitals have vital proposals for how well they use resources to create health policies. It is vital importance how public hospitals use resources efficiently to create health policies. Data Envelopment Analysis (DEA) which has with the same goals and objectives, especially, in measuring the productivity of businesses operating in the service sector is a special application of linear programming. Conventional DEA models can only be applied in cases where used input and produced outputs are strictly known. In order to be able to measure activity in cases where the data are not known precisely, Fuzzy Data Envelopment Analysis (FDEA) models was developed. In this study, the activities of 18 public hospitals which based on public hospitals administration of Samsun were measured. The data and literature obtained from the creation of the most appropriate input and output data sets for the efficiency measurement of hospitals were used. It was determined 4 inputs (Number of Doctors, Number of Outpatient Clinics, Total Number of Beds in Services, and Intensive Care Bed) and 4 outputs (Total number of examinations, operations, staying days, and inpatients) for activity measurement. Data related to the variables used in the research were officially obtained from the Ministry of Health, General Secretariat of the Public Hospitals Administration. Although the 12-month data between 2014 and 2017 were obtained for each year, It was evaluated annual periods. In order to determine the efficiency scores of hospitals (Decision Making Unit-DMU), both input and output oriented DEA models are used in frame of variable by scale (CCR) and fixed return on scale (BCC), in addition, in this study, the Saati- Memariani model which one of the fuzzy data enveloping (FDEA) models was used. In the results of study, number of examinations and person per doctor in Samsun is above the Turkey average. Besides, it was determined that there is a strong positive correlation between the input and output variables to be used for Data Envelopment Analysis (DEA) within the scope of the study. From the decision-making units (18 public hospitals) considered in the study, it has been found that a large part of the input and output focused CCR models (12 public hospitals) are effective. It was found that the activity values are partially increased from 2014 to 2017. According to the input-focused BCC model of Decision-Making Unit-DMU in the BCC model, it was found that 16 hospitals in 2014, 15 hospitals in 2015, 14 hospitals in 2016, and 15 hospitals in 2017 are active. Also, according to the output-focused BCC model, it was determined that 16 hospitals in 2014, 15 hospitals in 2015, 14 hospitals in 2016, and 15 hospitals in 2017 are active. It has been determined that activitiy changes of the evaluated decision-making units within the scope of the study was varied over the years. According to the results of Saati- Memariani model which one of the fuzzy data enveloping (FDMU) models to determinate the activities of decision-making units, according to the determined alpha cuts of each decision-making units (DMU) activity values for averages, lower and upper limit values are found to be different. According to the upper limit, which is the optimistic approach all DMU's were determined to be active.
Health services are directly affecting social life, especially people's life. The funding of hospitals based at the healthcare sector is provided by public or private sources. Public hospitals have vital proposals for how well they use resources to create health policies. It is vital importance how public hospitals use resources efficiently to create health policies. Data Envelopment Analysis (DEA) which has with the same goals and objectives, especially, in measuring the productivity of businesses operating in the service sector is a special application of linear programming. Conventional DEA models can only be applied in cases where used input and produced outputs are strictly known. In order to be able to measure activity in cases where the data are not known precisely, Fuzzy Data Envelopment Analysis (FDEA) models was developed. In this study, the activities of 18 public hospitals which based on public hospitals administration of Samsun were measured. The data and literature obtained from the creation of the most appropriate input and output data sets for the efficiency measurement of hospitals were used. It was determined 4 inputs (Number of Doctors, Number of Outpatient Clinics, Total Number of Beds in Services, and Intensive Care Bed) and 4 outputs (Total number of examinations, operations, staying days, and inpatients) for activity measurement. Data related to the variables used in the research were officially obtained from the Ministry of Health, General Secretariat of the Public Hospitals Administration. Although the 12-month data between 2014 and 2017 were obtained for each year, It was evaluated annual periods. In order to determine the efficiency scores of hospitals (Decision Making Unit-DMU), both input and output oriented DEA models are used in frame of variable by scale (CCR) and fixed return on scale (BCC), in addition, in this study, the Saati- Memariani model which one of the fuzzy data enveloping (FDEA) models was used. In the results of study, number of examinations and person per doctor in Samsun is above the Turkey average. Besides, it was determined that there is a strong positive correlation between the input and output variables to be used for Data Envelopment Analysis (DEA) within the scope of the study. From the decision-making units (18 public hospitals) considered in the study, it has been found that a large part of the input and output focused CCR models (12 public hospitals) are effective. It was found that the activity values are partially increased from 2014 to 2017. According to the input-focused BCC model of Decision-Making Unit-DMU in the BCC model, it was found that 16 hospitals in 2014, 15 hospitals in 2015, 14 hospitals in 2016, and 15 hospitals in 2017 are active. Also, according to the output-focused BCC model, it was determined that 16 hospitals in 2014, 15 hospitals in 2015, 14 hospitals in 2016, and 15 hospitals in 2017 are active. It has been determined that activitiy changes of the evaluated decision-making units within the scope of the study was varied over the years. According to the results of Saati- Memariani model which one of the fuzzy data enveloping (FDMU) models to determinate the activities of decision-making units, according to the determined alpha cuts of each decision-making units (DMU) activity values for averages, lower and upper limit values are found to be different. According to the upper limit, which is the optimistic approach all DMU's were determined to be active.
Description
Tez (yüksek lisans) -- Ondokuz Mayıs Üniversitesi, 2018
Libra Kayıt No: 124939
Libra Kayıt No: 124939
Citation
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
135
