The Prediction Of The Demand Of Nursing Assistants In China Based On Markov Model

ObjectiveThis paper tries to forecast the population size of the elderly who need nursing care and the scale of nursing assistants,which serves as a reference for related bodies to formulate medium-and-long term plan on the training of nursing assistants,thus bettering nursing quality and meeting the requirement of the fast aging process of the population.MethodsData employed in this study are from four sources.Source One is the longitudinal survey data of “China Health and Retirement Longitudinal Study”(CHARLS)in 2011,2013 and 2015 released by the Center for Healthy Aging and Development Studies,Peking University.Based on the survey data of CHARLS in 2011 and 2013,this study estimates the transition probability of the health conditions of the elderly with Markov condition-transition estimation method.Source Two is the 6th National Population Census in 2010.Health conditions of the population aged 45 and above is estimated based on data in CHARLS 2011 and the population sizes of different conditions in 2010 are thus estimated.Based on the transition probability got from Resource One,population(aged 65 and above)sizes of different conditions by 2030 are estimated.Source Three is the required ratio of nursing assistant to the elderly in nursing institutions set by Regulations on Nursing Institutions for the Elderly in several cities and provinces.By determining the required ratio,the demand for nursing assistants by 2030 in China is estimated.Source Four is data from the CHARLS 2015.By comparing the estimated results with the data in CHARLS 2015 released lately,the validity of Markov approach to estimating the need for nursing staff is justified.ResultsThe sample size of this study is 32713: 14911 are aged 45 and above and surveyed by both CHARLS 2011 and 2013,and 17802 are aged 49 and above and surveyed by CHARLS 2015.Health conditions of study objects from CHARLS 2011 and 2013 in 2011 are as following: 314(2.1%)are in need of device aid and 2242(15%)are in need of nursing;while in 2013,11733(78.7%)do not need care,390(2.6%)are in need of device aid,2401(16.1%)are in need of nursing,and 387(2.6%)died.Among study objects from 2011 to 2013,10650 keep needing no care,242 become in need of device aid from self-caring,1281 become in need of nursing from self-caring,and 182 self-cared ones died;133 become self-cared from in need of device aid,52 keep in need of device aid,103 become in need of nursing from in need of device aid,and 26 device-aided ones died;950 become self-cared from in need of nursing,96 become in need of device aid from in need of nursing,1017 become in need of nursing from in need of device aid,and 179 nursed ones died.As of health conditions of study objects in CHARLS 2015,14284(80.2%)do not need care,3209(2.6%)are in need of device aid,and 3053(17.1%)are in need of nursing.Estimation results for transition probability of health conditions in population aged 49 and above in 2015 are that 336 million keep needing no care,11.8 million are in need of device aid and 77.4 million are in need of nursing.Comparison by ?2 test between estimation results and actual conditions shows no difference: ?2=5.309,P>0.05,i.e.Logit model and Markov model can be used in the study of health conditions in the middle-aged and elderly population.It is forecast that by 2030,among the elderly aged 65 and above,there will be 280 million self-cared ones,10 million in need of device aid and 67 million in need of nursing.The total demand for nursing assistants would be 50 million by 2030.ConclusionsThis paper indicates that in 2011,among study subjects in the same health condition,the ones aged younger show a higher probability of keeping or recovering back to being self-cared as well as a lower mortality rate than those aged older.Transition probabilities of health conditions for the elderly at different ages show difference but the same variation tendency.The worse the health condition is,the harder it will be to recover back to being self-cared or better health conditions.And the risk of mortality becomes higher with worsening health conditions.Markov model is applicable to forecast the population size of the elderly and the demand for nursing assistants.Moreover,the forecasting precision is high after validation.The gap between the forecasting results of this study and the actual size of China's nursing assistants demonstrates severe imbalance between the supply and demand of nursing staff.Therefore,it is safe to draw the conclusion that reasonable planning and training of nursing assistants is of significance.