This story was originally published on Feb. 9, 2013

TORONTO - Following a W5 investigation into abuse in long-term care facilities in Ontario that aired in March 2012, we received many emails and phone calls from across Canada telling us of similar cases.

Looking for national statistics, we discovered none are kept. We decided to compile our own.

We identified 38 jurisdictions that kept records of abuse in long-term care: six provincial ministries and in four provinces, 32 regional health authorities.

The next step was to file access to information (ATI) requests with each. Our first requests were made in June 2012, and soon thousands of pages of documents began arriving.

We had initially sought occurrence reports for each reported incident but government agencies refused to provide this data, citing privacy concerns, or sought huge fees for searching, redacting private information and copying.

The estimate ran to about $170,000. However, if we were prepared to accept just simple statistics the cost was more reasonable, $1,000 dollars in total.

In the end, we received a mix of statistics and, in some cases, occurrence reports. The next challenge was to analyze the data to produce an estimate of the total resident-on-resident incidents in one year.

We enlisted Prof. Lynn McDonald and her team of researchers at the University of Toronto's Institute for Life Course and Aging.

"In Canada we've never had a study on abuse in any institution, let alone on resident to resident," McDonald told W5.

The documents obtained under the ATI requests showed 23,521 occurrences of all types in one year. In 26 of the 38 jurisdictions, the data provided specific information about the type of incident – 6,494 resident-on-resident altercations.

The challenge was to calculate numbers for Quebec, Alberta, Prince Edward Island and nine health authorities that provided a total number of incidents, but without differentiation.

McDonald's team looked at our data and pointed out that among the clearly differentiated cases, the rate of resident-on-resident occurrences was 43 per cent.

Applied to the total numbers of all incidents in the 12 jurisdictions for which we did not receive specific data, this implied a further 3,553 resident-on-resident confrontations, for a total single-year estimate of 10,047 – an astounding national number and one never before reported.

And given the culture of underreporting that is prevalent in the long-term sector, the real number is likely higher.

Just how much underreporting there may be can be seen in two sets of data we received from the Mamawetan Churchill River Health Authority in Saskatchewan. With staff using normal methods of providing data, the total number of incidents in 2011 was reported to be 23.

But the health authority conducted a "Violence Identification Audit" during two weeks in September 2011. It's a program that allows staff to anonymously report incidents.

During that two week period, in a single home with 16 beds, staff reported 89 total incidents – way more than the "official" record provided to W5.

One of the immediate outcomes of working with W5 and analyzing the data collected through ATI requests is that McDonald and her team will continue the research begun "on our own accord, to dig further into the data and collect more in order to better understand the scope of elder abuse in Canada."

"When CTV came to me I thought 'Oh my goodness, this is the most data I've ever seen on this particular issue.' And when you really look at it, it still has a number of places where there's holes that we need more information. We need to know what we're doing," McDonald said.