The paper contributes on understanding the most performing category of hospitals and operation in relation to the size (scale efficiency) taking the case of private hospital in Tanzania. Using the sample of 34 private not for profit (PNFP) hospitals, the effect of hospitals size on efficiency was investigated. Data were extracted from respective annual hospital’s report from 2009 -2013. Data Envelopment Analysis (DEA) model was employed to compute efficiency based on the hospitals size. Findings revealed an average of variable return to scale technical efficiency (VRSTE) scores for the large hospitals is 98.8%, meaning hospitals could have produced 1.2% more outputs with the same volume of inputs. Average VRSTE score for medium and small hospitals is 79.45% and 88.6% respectively. Large private hospitals are more efficient compared to their medium and small counterparts. 77.7% of large private hospitals were efficient, meanwhile 47.05% and 62.5% of medium and small hospitals respectively were found to be efficient. However, the average scale efficiency of large hospitals is 91.2 %. Further, medium and small private hospitals could reduce their sizes by 19.05% and 3.5% respectively to become scale efficient. The study recommend medium hospitals with increasing return to scale (IRS) should increase scale of their activities to enjoy economies of scale while the hospitals experiencing decreasing return to scale (DRS) should reduce the inputs to avoid diseconomies of scale. Policy makers should focus on modeling the size of the hospitals against their respective scale of activities. Improvement in efficiency based on the hospitals size will prevent loss of scarce hospitals resources caused by scale inefficiency.