Describe the health information system and the uses of medical records.

Health Information System and Records

• Its components and sources — the census; civil registration of births and deaths; the sample registration system; notification of diseases; hospital and clinic records; disease registers such as cancer registries; surveys, notably the national family health survey; epidemiological surveillance such as IDSP; environmental and health manpower statistics; and records of health expenditure
• The requirements of a good system — the data must be relevant to a decision that will actually be taken; accurate and complete; timely, since information arriving after the decision is worthless; simple to collect; standardised and comparable; confidential; and actually used
• The INDIAN systems — the health management information system (HMIS), a web portal for facility-level service data; the reproductive and child health portal, which is name-based and tracks individual women and children; nikshay for tuberculosis; IDSP and the Integrated Health Information Platform; and Co-win and U-win for immunisation
• Name-based tracking is a genuine advance over aggregate counting, since it allows a specific woman who has missed a visit to be followed up, which no aggregate report can ever do
• The civil registration system records births and deaths under the Registration of Births and Deaths Act, 1969, which makes registration compulsory
• Birth registration is now nearly complete, but medical certification of the cause of death remains poor, particularly for deaths occurring at home, which are the majority in rural India
• The consequence is that India knows fairly well how many people die and much less well what they die OF, which limits every attempt to set priorities by cause of death
• Verbal autopsy is the partial remedy — a structured interview with relatives after a home death, from which a probable cause is assigned; it is imprecise for individual cases but useful for population-level patterns
• The sample registration system (SRS) provides annual estimates of birth rate, death rate, infant mortality and total fertility for states
• Its method is dual record — continuous enumeration by a resident recorder combined with an independent half-yearly survey, with the two lists matched and discrepancies field-verified
• The dual record design is what gives the SRS its accuracy, since each method catches events the other misses, and the matching measures the omission rather than merely hoping it is small
• The international classification of diseases (ICD) provides the standard coding that makes data comparable between places and over time
• Medical records — their purposes — continuity of patient care, which is the primary purpose; communication between providers; legal evidence; medico-legal and insurance requirements; audit and quality improvement; research and teaching; and planning and statistics
• The characteristics of a good record — accurate, legible, complete, dated and signed, contemporaneous, objective, and confidential
• Records should never be altered after the event; a correction should be made as a dated addition with the original remaining legible, since an altered record destroys the credibility of the whole document
• The common problems of the Indian system — over-collection of data that nobody analyses; the ANM spending a large share of her time on registers rather than on people; duplicate registers for different vertical programmes; data reported upward but never fed back downward; inflation of figures to meet targets; near-total exclusion of the private sector; and delay
• The absence of feedback is the most corrosive failing — a worker who never sees any use made of the data she collects will not take care over collecting it, so feedback is a means of improving data quality and not merely a courtesy
• India knows how many die better than what of, since certification is poor for the home deaths that predominate rurally
• Feedback is a data quality measure, since a worker who never sees the data used will not collect it carefully
• Name-based tracking allows individual follow-up, which no aggregate report can ever do
• The dual record system measures its own omission rather than hoping it is small, which is what makes SRS accurate
• Verbal autopsy suits populations, not individuals, being imprecise per case but useful in aggregate
• Registration of births and deaths is compulsory under the 1969 Act, and birth registration is now nearly complete
• Never alter a record after the event; add a dated correction and leave the original legible
• An altered record discredits the whole document, which matters medico-legally far more than the original error would have
• The ANM spends much of her time on registers, which is time not spent on the people the registers describe
• Duplicate registers follow vertical programmes, so integration of records lags behind integration of services
• The private sector is nearly absent from the data, which distorts every estimate built on facility reporting
• Information arriving after the decision is worthless, which makes timeliness a requirement rather than a preference
• ICD coding makes data comparable between places and across time, which is its entire purpose