Chinese Medical Journal

ORIGINAL ARTICLE
Year
: 2018  |  Volume : 131  |  Issue : 4  |  Page : 389--394

Cost Analysis of Cervical Cancer Patients with Different Medical Payment Modes Based on Gamma Model within a Grade A Tertiary Hospital


Suo-Wei Wu, Tong Chen, Qi Pan, Liang-Yu Wei, Qin Wang, Jing-Chen Song, Chao Li, Ji Luo 
 Department of Medical Administration, Beijing Hospital, National Center of Gerontology, Beijing 100730, China

Correspondence Address:
Dr. Qi Pan
Department of Medical Administration, Beijing Hospital, National Center of Gerontology, Beijing 100730
China

Abstract

Background: Cervical cancer shows a growing incidence and medical cost in recent years that has increased severe financial pressure on patients and medical insurance institutions. This study aimed to investigate the medical economic characteristics of cervical cancer patients with different payment modes within a Grade A tertiary hospital to provide evidence and suggestions for inpatient cost control and to verify the application of Gamma model in medical cost analysis. Methods: The basic and cost information of cervical cancer cases within a Grade A tertiary hospital in the year 2011–2016 were collected. The Gamma model was adopted to analyze the differences in each cost item between medical insured patient and uninsured patients. Meanwhile, the marginal means of different cost items were calculated to estimate the influence of payment modes toward different medical cost items among cervical cancer patients in the study. Results: A total of 1321 inpatients with cervical cancer between the 2011 and 2016 were collected through the medical records system. Of the 1321 cases, 65.9% accounted for medical insured patients and 34.1% were uninsured patients. The total inpatient medical expenditure of insured patients was RMB 29,509.1 Yuan and uninsured patients was RMB 22,114.3 Yuan, respectively. Payment modes, therapeutic options as well as the recurrence and metastasis of tumor toward the inpatient medical expenditures between the two groups were statistically significant. To the specifics, drug costs accounted for 37.7% and 33.8% of the total, surgery costs accounted for 21.5% and 25.5%, treatment costs accounted for 18.7% and 16.4%, whereas the costs of imaging and laboratory examinations accounted for 16.4% and 15.2% for the insured patient and uninsured patients, respectively. As the effects of covariates were controlled, the total hospitalization costs, drug costs, treatment costs as well as imaging and laboratory examination costs showed statistical significance. The total hospitalization costs, drug costs, treatment costs as well as imaging and laboratory examination costs of insured patient were 1.33, 1.42, 1.52, and 1.44 times of uninsured patients. Conclusions: The analysis of different payment modes toward the medical economic characteristics based on Gamma model is basically rational. Medical payment modes are having certain influence toward the hospitalization expenses of cervical cancer patients in an extent, as drug costs, treatment costs, and examination costs appear to be the main causes.



How to cite this article:
Wu SW, Chen T, Pan Q, Wei LY, Wang Q, Song JC, Li C, Luo J. Cost Analysis of Cervical Cancer Patients with Different Medical Payment Modes Based on Gamma Model within a Grade A Tertiary Hospital.Chin Med J 2018;131:389-394


How to cite this URL:
Wu SW, Chen T, Pan Q, Wei LY, Wang Q, Song JC, Li C, Luo J. Cost Analysis of Cervical Cancer Patients with Different Medical Payment Modes Based on Gamma Model within a Grade A Tertiary Hospital. Chin Med J [serial online] 2018 [cited 2018 Nov 16 ];131:389-394
Available from: http://www.cmj.org/text.asp?2018/131/4/389/225052


Full Text



 Introduction



As one of the most commonly seen gynecologic malignant tumors, cervical cancer is showing a growing tendency of incidence and medical cost in recent years.[1] The treatment of cervical cancer is mainly based on surgeries combined with radiotherapy and chemotherapy that usually takes a long period of time, and the costs of therapies are growing year on year. As one of the medical insurance coverage disease types, its long therapeutic cycle and high treatment costs are bringing severe financial pressures to the patients and medical insurance institutions that have gradually arouse the attention of health administrative departments. The reasonable cost control of inpatient medical expenditures for cervical cancer is crucial in easing the economic burden of patients and medical insurance reimbursements.[2] Extensive studies showed that among varied factors that affect the hospitalization expenses, payment modes can be counted as one of the most important aspects since medical insurances may compensate a great proportion of medical costs that are likely to increase the possibility of overtreatment despite the fact in reducing the financial burden of patients.[3]

As a type of sophisticated statistical model, gamma distribution has been widely applied in cost analysis in recent years due to its universal characteristics in the research of health economics.[4] It is believed that logarithmic linked Gamma model is more suitable for cost analysis since the estimated values are closer to the actual value in compare with multiple linear regression method according to relevant researches. In view of the skewed distribution of the cost, the Gamma model is able to process the data with log linked gamma distribution to ensure the integrity of original information and avoid conversion bias.[5] However, extensive studies of medical cost analysis using Gamma model are still rare. This paper chose the inpatient cervical cancer cases within a Grade A tertiary hospital of the year 2011–2016 as research samples to verify the application of Gamma model in medical cost analyses. Furthermore, it is hoped that the study of different medical payment modes toward hospitalization expenses can provide solid basis and rational suggestions of cost control for hospitals as well as health administrative and medical insurance departments.

 Methods



Acquisition and reduction of data and materials

The data and information of patients whose main diagnoses are labeled as cervical cancer from January 1, 2011, to December 31, 2016, within a Grade A tertiary hospital of Beijing were collected through the medical record management system. The data and materials collected included the general situation of the patients (such as age, gender, operations, payment modes, length of hospital stays, admission conditions, treatments, and follow-up situations) and medical costs information including total costs, bed charges, drug costs, treatment costs, examination costs, and other costs.[6] In regards of eliminating the impact of price fluctuation toward the study to realize the comparability of time span,[7] the costs of hospitalization were adjusted according to consumer price index in the “China statistical yearbook – Category of health” of the year 2011–2016, with 2016 as the base year [Table 1].{Table 1}

Statistical analysis

The basic information of patients was examined through Kolmogorov-Smirnov normality test.[8] Since age, length of hospital stay, and medical costs were not subject to normal distribution, median numbers were adopted in the description, while percentages were adopted to indicate the constitution of each cost item toward the total, and a relative number was used in describing categorical variables.[9] Meanwhile, Chi-square test was used to analyze the difference between the basic information of medical insured and uninsured patients. Setting each medical cost item as dependent variables and set age, marriage status, payment modes, therapeutic options, surgical treatments, pathological type, lymphatic metastasis, as well as distant metastasis and recurrence of the patients as independent variables to construct the logarithmic linked Gamma model.[10] Analyze the variance of inpatient medical expenditures of patients with and without medical insurance as well as the influencing factors of hospitalization expenses. Meanwhile, marginal costs of medical insured patients and uninsured patients were estimated in the study.[11] All the data and material collected were entered into Excel 2010 software for Microsoft (Microsoft Corporation, Washington, USA), and statistical analyses were performed using SPSS software version 21.0 (SPSS Inc., Chicago, IL, USA).

 Results



General situations

A total of 1321 cases of cervical cancer were enrolled in this study, among which 65.9% were medical insured patients and 34.1% were uninsured patients. The median length of hospital stay was 10.5 days and the median age of the enrolled cases was 49.3 years, and 79.9% of patients were performed with surgeries. The general situations of medical insured patients and uninsured patients were shown in [Table 2], in which the marital status and treatment options were statistically significant, while other factors showed no significant difference between the two groups.{Table 2}

Inpatient medical expenditure of insured and uninsured patients

In the study, the inpatient medical expenditures of medical insured and uninsured patients were RMB 29,509.1 Yuan and RMB 22,114.3 Yuan, respectively. Of the specifics, drug costs, surgery costs as well as treatment costs were the top three cost categories, with drug costs accounted for 37.7% and 33.8% of the total costs, surgery costs accounted for accounted for 21.5% and 25.5%, treatment costs accounted for 18.7% and 16.4%, whereas the costs of imaging and laboratory examinations accounted for 16.4% and 15.2% among insured and uninsured patients, respectively, as shown in [Table 3].{Table 3}

Influence of payment modes toward inpatient medical expenditures of cervical cancer patients

The results of Gamma analysis between different payment modes toward the inpatient medical expenditures of cervical cancer were shown in [Table 4]; as payment modes, therapeutic options as well as the recurrence and metastasis of tumor were counted as statistically significant. Further analyses on medical costs of the insured and uninsured inpatient costs as we regulated the covariates were shown in [Table 5], while total costs, drug costs, treatment costs as well as examination costs were significantly different between medical insured and uninsured patients, as the total hospitalization costs, drug costs, treatment costs, and examination costs of insured inpatient were 1.33, 1.42, 1.52, and 1.44 times of uninsured patients, respectively.{Table 4}{Table 5}

 Discussion



Influence of payment modes toward total inpatient medical expenditures among cervical cancer patients

As is widely known, payment modes have significant influence toward medical expenditures in clinical. According to the study, marital status and treatment options were statistically significant in inpatient medical expenditures between medical insured patient and uninsured patients.[12] Notably, marital status is somehow related to the medical insurance type among some patients and that medical insurance status may influence patients' choices on treatment options.[13] Thus, to exclude the influence of covariates, Gamma model was adopted to control the other covariates in further analyses, and results showed that the hospitalization expenses among cervical cancer patients with medical insurance were significantly higher than those without, this is in accordance with other domestic and foreign researches.[14] Here are several possible reasons:first, the proportion of uninsured patients in China is still quite large since the medical insurance coverage is not complete, which may place limits to their medical behavior in a certain degree when seeking medical treatments.[15] Second, according to relevant studies, different medical payment modes are having certain stimulations and inducements toward doctors in their therapeutic behaviors. This may largely attribute to the self-financing reality of Chinese hospitals since health-care funds in China are seriously insufficient, and that some medical service providers are inclined to provide excessive treatments due to economic-induced demands since it is directly related to the profit of the hospitals and the income of medical staffs.[16] Last but not least, the diagnosis of cervical cancer was selected as the trial item in thefirst diagnosis-related groups' prospective payment system categories in China, which might result to excessive medical services among patients with health-care insurance.[17]

Influence of therapeutic options as well as recurrence and metastasis of tumor toward total inpatient medical expenditures among cervical cancer patients

According to Gamma model analysis, the results of therapeutic options as well as the recurrence and metastasis of tumor toward the inpatient medical expenditures of cervical cancer were statistically significant. In regard to the characteristics of cervical cancer, it was explicable that the medical costs of different therapeutic options vary significantly since a great amount of adjunctive therapies and maintenance therapy (such as chemotherapies, radiotherapy as well as immunotherapies) might be involved in the treatment. While the recurrence and metastasis of tumor reflect the severity of disease, the status of patients, and the difficulty in further treatment, it is obvious that the variance in the indicator has significant differences toward the overall medical costs among cervical cancer patients.

Influence of payment modes toward different medical cost items among cervical cancer patients

In the research, Gamma model was also applied in analyzing the average medical expenses toward different cost items in the control of other covariates. Moreover, results showed that drug costs, treatment costs, and examination costs of medical insured patients are 1.42, 1.52, and 1.44 times of uninsured patients. On the contrary, since other costs items (such as surgery costs, bed charges as well as nursing costs) are nonnegligible and irreplaceable for the treatment, no significant difference between medical insured and uninsured patients was shown in the study. Furthermore, it is believed that the differences in drug costs will gradually be narrowed between medical insured and uninsured patients, and the gaps of other cost items will be relatively expanded with the implementation of health-care reform policies when the profits of drug makeups were cancelled, and the usage of unnecessary drugs was further cut down by the hospital.[18]

Application of Gamma model in medical cost analysis

The statistical methods that most frequently used in medical cost analysis including multiple linear models, logistic regression as well as various multivariate statistical analyses.[19] As for the cost analyses of nonnormal distributions, the most commonly used method is the construction of the multiple linear model by the logarithm transformation of the original data. However, the application of the method calls for strict conditions, and the intercept estimation bias may exist in the predictive values of exponentiation cost estimations.[20] Logistic regression does not require normal distribution and homogeneity of variance, but the integrity of data information was reduced through the classification of the continuous cost data.[21] In comparison, the Gamma model processes the data with the log linked gamma distribution with regard to the skewed distribution characteristics of the cost data, ensures the integrity of the original data information, and avoids the conversion bias.[22] The results of similar studies show that the Gamma model is more suitable for a medical cost analysis by comparing the results of multiple linear regressions and the logarithmic link Gamma model since the estimated average cost values through Gamma model are closer to the actual value.[23] Thus, the analysis of medical expenditures based on Gamma model could be applied to other similar single diseases such as uterine benign diseases, normal deliveries, and other common diagnoses of gynecology.

Since the study was based on a single-centered hospital, considering the differences in economic level, medical insurance policies as well as epidemiology of the disease, the generalization of the results should be more careful in other regions and circumstances.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM, et al. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127:2893-917. doi: 10.1002/ijc.25516.
2Singh E, Seth S, Rani V, Srivastava DK. Awareness of cervical cancer screening among nursing staff in a tertiary institution of rural India. J Gynecol Oncol 2012;23:141-6. doi: 10.3802/jgo.2012.23.3.141.
3Krishnan S, Madsen E, Porterfield D, Varghese B. Advancing cervical cancer prevention in India: Implementation science priorities. Oncologist 2013;18:1285-97. doi: 10.1634/theoncologist.18-S2-13.
4Sankaranarayanan R, Esmy PO, Rajkumar R, Muwonge R, Swaminathan R, Shanthakumari S, et al. Effect of visual screening on cervical cancer incidence and mortality in Tamil Nadu, India: A cluster-randomised trial. Lancet 2007;370:398-406. doi: 10.1016/S0140-6736(07)61195-7.
5Ince EZ, Cekmez F, Yıldırım Ş, Demirel A, Bilgic B, Kılıcaslan I, et al. Malignant epignathus including a nephroblastoma component and successful management. Ann Diagn Pathol 2013;17:288-90. doi: 10.1016/j.anndiagpath.2012.01.008.
6Singh M, Ranjan R, Das B, Gupta K. Knowledge, attitude and practice of cervical cancer screening in women visiting a tertiary care hospital of Delhi. Indian J Cancer 2014;51:319-23. doi: 10.4103/0019-509X.146780.
7Liu X, Zhang W. Effect of maternal age on pregnancy: A retrospective cohort study. Chin Med J 2014;127:2241-6. doi: 10.3760/cma.j.issn.0366-6999.20133118.
8Schiffman M, Wentzensen N. Human papillomavirus infection and the multistage carcinogenesis of cervical cancer. Cancer Epidemiol Biomarkers Prev 2013;22:553-60. doi: 10.1158/1055-9965.EPI-12-1406.
9Bacci S, Bartolucci F, Minelli L, Chiavarini M. Preterm birth: Analysis of longitudinal data on siblings based on random-effects logit models. Front Public Health 2016;4:278. doi: 10.3389/fpubh.2016.00278.
10Sindhuphak R, Issaravanich S, Udomprasertgul V, Srisookho P, Warakamin S, Sindhuphak S, et al. Anew approach for the detection of cervical cancer in Thai women. Gynecol Oncol 2003;90:10-4. doi: 10.1016/S0090-8258(03)00196-3.
11Rubina S, Vidyasagar M, Krishna CM. Raman spectroscopic study on prediction of treatment response in cervical cancers. J Innov Opt Health Sci 2013;6:115-21. doi: 10.1142/S1793545813500144.
12Sandeva M, Uchikova E. Frequency and medical social aspects premature birth (in Bulgarian). Akush Ginekol (Sofiia) 2016;55:27-33.
13Sankaranarayanan R, Ferlay J. Worldwide burden of gynaecological cancer: The size of the problem. Best Pract Res Clin Obstet Gynaecol 2006;20:207-25. doi: 10.1016/j.bpobgyn.2005.10.007.
14Xavier-Júnior JC, Dufloth RM, do Vale DB, Tavares TA, Zeferino LC. High-grade squamous intraepithelial lesions in pregnant and non-pregnant women. Eur J Obstet Gynecol Reprod Biol 2014;175:103-6. doi: 10.1016/j.ejogrb.2014.01.018.
15Quinn JA, Munoz FM, Gonik B, Frau L, Cutland C, Mallett-Moore T, et al. Preterm birth: Case definition and guidelines for data collection, analysis, and presentation of immunisation safety data. Vaccine 2016;34:6047-56. doi: 10.1016/j.vaccine.2016.03.045.
16Simoens S, Dunselman G, Dirksen C, Hummelshoj L, Bokor A, Brandes I, et al. The burden of endometriosis: Costs and quality of life of women with endometriosis and treated in referral centres. Hum Reprod 2012;27:1292-9. doi: 10.1093/humrep/des073.
17Li QJ, Cai JQ, Liu CY. Evolving molecular genetics of glioblastoma. Chin Med J 2016;129:464-71. doi: 10.4103/0366-6999.176065.
18Riazi H, Tehranian N, Ziaei S, Mohammadi E, Hajizadeh E, Montazeri A, et al. Clinical diagnosis of pelvic endometriosis: A scoping review. BMC Womens Health 2015;15:39. doi: 10.1186/s12905-015-0196-z.
19Qu YJ, Song F, Yang YL, Jin YW, Bai JL. Compound heterozygous mutation in two unrelated cases of Chinese spinal muscular atrophy patients. Chin Med J 2011;124:385-9. doi: 10.3760/cma.j.issn.0366-6999.2011.03.012.
20Wang BC, Ma J. Role of microRNAs in malignant glioma. Chin Med J 2015;128:1238-44. doi: 10.4103/0366-6999.156141.
21Victora CG, Huttly SR, Fuchs SC, Olinto MT. The role of conceptual frameworks in epidemiological analysis: A hierarchical approach. Int J Epidemiol 1997;26:224-7. doi: 10.1093/jie/26.1.224.
22Buck Louis GM, Hediger ML, Peterson CM, Croughan M, Sundaram R, Stanford J, et al. Incidence of endometriosis by study population and diagnostic method: The ENDO study. Fertil Steril 2011;96:360-5. doi: 10.1016/j.fertnstert.2011.05.087.
23Derakhshi B, Esmailnasab N, Ghaderi E, Hemmatpour S. Risk factor of preterm labor in the West of Iran: A case-control study. Iran J Public Health 2014;43:499-506.