Introduction

These data were downloaded through the Iris Quebec system. The first request of this data made sometime in the Fall of 2008. Unfortunately, data on the Iris Quebec system expires after a certain amount of time, and all of the data were not downloaded in time. This resulted in only Etude 1 being loaded into the IrisQuebec database on SARS. A copy of these data can be found on disk2/IrisQuebecData, and on \\tuberculosis\Projects\Data Sets\IRIS Quebec Downloads\First Run Data.

Separately, a request was made to RAMQ to find the FSA of each person in Etude 1 and 2. These data are stored as etude1_20070618a and etude2_20070618a in the IrisQuebec database on SARS.

A second request of data was made through the Iris Quebec system in order to collect the Etude 2 data and Khaled's data as well. The Etude 1 data was also re-requested. The results of this query are stored in \\tuberculosis\Projects\Data Sets\IRIS Quebec Downloads\ as well as on disk2/IrisQuebecData2 on SARS.

As of 2009-04-23, the Etude 1 first query data are available under the tables suffixed with _bak in the IrisQuebec database on SARS. These tables have been removed.

Another request had to made to RAMQ for the FSA info for the second request. (The encrypted NAMs changed, so we couldn't just use the first one.) These data are not in as of 2009-04-23. The etude1_20070618a and etude2_20070618a tables are from the first request and are therefore unusable with the second request data. These will be removed when the new data arrives.

The data had to be re-extracted a third time, for reasons which I can't remember.

The data had to be re-extracted a fourth time, because Jimmy identified a problem in the data where private clinics were not being included. The RAMQ rapidly provided the fourth dataset directly (without going through the IRIS Quebec system).

The fourth extraction still had problems with the place of visit, and the type of place of visit. More details can be found under Missing Place in Etude 1.

Inclusion Criteria

  • lived in Montreal during at least one point during the 11 year study period
  • had at least one ILI claim during the 11 year study period

For individuals who fulfilled these criteria, we then obtained ALL their data during the study period, irregardless of the diagnosis or whether they were living in Montreal on the date of service

Need to clarify for Etude 1:

  • we're not sure whether the conditions for the inclusion criteria had to occur simultaneously (i.e. during the same year)

Specific Inclusion Criteria

At any time between 1996 and 2006, the FSA of the place of residence must be in:

H1A, H1B, H1C, H1E, H1G, H1H, H1J, H1K, H1L, H1M, H1N, H1P, H1R, H1S, H1T, H1V, H1W, H1X, H1Y, H1Z, H2A, H2B, H2C, H2E, H2G, H2H, H2J, H2K, H2L, H2M, H2N, H2P, H2R, H2S, H2T, H2V, H2W, H2X, H2Y, H2Z, H3A, H3B, H3C, H3E, H3G, H3H, H3J, H3K, H3L, H3M, H3N, H3P, H3R, H3S, H3T, H3V, H3W, H3X, H3Y, H3Z, H4A, H4B, H4C, H4E, H4G, H4H, H4J, H4K, H4L, H4M, H4N, H4P, H4R, H4S, H4T, H4V, H4W, H4X, H4Y, H4Z, H5A, H5B, H7A, H7B, H7C, H7E, H7G, H7H, H7J, H7K, H7L, H7M, H7N, H7P, H7R, H7S, H7T, H7V, H7W, H7X, H7Y, H8N, H8P, H8R, H8S, H8T, H8Y, H8Z, H9A, H9B, H9C, H9E, H9G, H9H, H9J, H9K, H9P, H9R, H9S, H9W, H9X, J0K, J0L, J0N, J0P, J0R, J0S, J0V, J3E, J3G, J3H, J3L, J3N, J3V, J3X, J3Y, J3Z, J4B, J4G, J4H, J4J, J4K, J4L, J4M, J4N, J4P, J4R, J4S, J4T, J4V, J4W, J4X, J4Y, J4Z, J5A, J5B, J5C, J5K, J5L, J5M, J5R, J5T, J5W, J5X, J5Y, J5Z, J6A, J6J, J6K, J6N, J6R, J6V, J6W, J6X, J6Y, J6Z, J7A, J7B, J7C, J7E, J7G, J7H, J7J, J7K, J7L, J7M, J7N, J7P, J7R, J7T, J7V, J7W, J7X, J7Y, J7Z, J8H, J8L

Extent of Etude 1 FSAs

For reasons I do not understand, I cannot find the geographical extent of these five FSAs: H4Y, H4Z, H5A, H5B, J7W. All five of these FSAs definitely exist in 2010 as well as some time during 1996-2006. They all have postal codes in the PCCF. However, interestingly, some of them (H4Z, H5A, H5B) show up on the FSA suppression list.

Additionally, you may note that several areas in the northwest are not contiguous with the rest of the FSA set. J0V is a non-contiguous FSA, and is responsible for most of the detached portion in the northwest.

The inclusion of J8L, which is in the far west side, is inexplicable.

AND they must have either one of these RAMQ ICD9 codes:

010, 0100, 0101, 0108, 0109, 011, 0110, 0111, 0112, 0113, 0114, 0115, 0116, 0117, 0118, 0119, 020, 0203, 0204, 0205, 0219, 022, 0221, 024, 0249, 025, 0259, 032, 0320, 0321, 0322, 0323, 0329, 033, 0330, 0331, 0338, 0339, 034, 0340, 0529, 0551, 073, 0739, 0741, 0790, 0793, 0798, 0799, 0830, 1124, 1149, 115, 1150, 1151, 1159, 1309, 1363, 382, 3820, 3824, 3829, 460, 4609, 461, 4618, 4619, 462, 4629, 463, 4639, 464, 4640, 4641, 4642, 4643, 4644, 465, 4650, 4658, 4659, 466, 4660, 4661, 4789, 480, 4800, 4801, 4802, 4808, 4809, 481, 4819, 482, 4820, 4821, 4822, 4823, 4824, 4828, 4829, 483, 4839, 484, 4841, 4843, 4845, 4846, 4847, 4848, 485, 4859, 486, 4869, 487, 4870, 4871, 4878, 490, 4909, 491, 4910, 4911, 4918, 4919, 507, 5070, 5071, 5078, 511, 5110, 5111, 5118, 5119, 513, 5130, 5131, 518, 5180, 5184, 5188, 5192, 7806, 7841, 7860, 7861, 7862, 7865, 7953, V018

AND / OR one of these MEDECHO codes

0100, 0101, 0108, 0109, 0110, 0111, 0112, 0113, 0114, 0115, 0116, 0117, 0118, 0119, 0203, 0204, 0205, 0219, 0221, 0249, 0259, 0320, 0321, 0322, 0323, 0329, 0330, 0331, 0338, 0339, 0340, 0529, 0551, 0739, 0741, 0790, 0793, 0798, 0799, 0830, 1124, 1149, 1150, 1151, 1159, 1309, 1363, 3820, 3824, 3829, 4609, 4618, 4619, 4629, 4639, 4640, 4641, 4642, 4643, 4644, 4650, 4658, 4659, 4660, 4661, 4789, 4800, 4801, 4802, 4808, 4809, 4819, 4820, 4821, 4822, 4823, 4824, 4828, 4829, 4839, 4841, 4843, 4845, 4846, 4847, 4848, 4859, 4869, 4870, 4871, 4878, 4909, 4910, 4911, 4918, 4919, 5070, 5071, 5078, 5110, 5111, 5118, 5119, 5130, 5131, 5180, 5184, 5188, 5192, 7806, 7841, 7860, 7861, 7862, 7865, 7953, V018, 01000, 01001, 01002, 01003, 01004, 01005, 01009, 01010, 01011, 01012, 01013, 01014, 01015, 01019, 01080, 01081, 01082, 01083, 01084, 01085, 01089, 01090, 01091, 01092, 01093, 01094, 01095, 01099, 01100, 01101, 01102, 01103, 01104, 01105, 01109, 01110, 01111, 01112, 01113, 01114, 01115, 01119, 01120, 01121, 01122, 01123, 01124, 01125, 01129, 01130, 01131, 01132, 01133, 01134, 01135, 01139, 01140, 01141, 01142, 01143, 01144, 01145, 01149, 01150, 01151, 01152, 01153, 01154, 01155, 01159, 01160, 01161, 01162, 01163, 01164, 01165, 01169, 01170, 01171, 01172, 01173, 01174, 01175, 01179, 01180, 01181, 01182, 01183, 01184, 01185, 01189, 01190, 01191, 01192, 01193, 01194, 01195, 01199, 05290, 05291, 05297, 05298, 05299, 51881, 51882, 51889, 78650, 78651, 78652, 78659

is included

Differences in the Second Data Request

The data in second request were not exactly the same as the first request. Two new tables were added:

  • Modificateurs
  • Secteur d'activité.

One table was renamed: *Code du médicament (DIN) renamed to Numéro d'identification de médicament (DIN).

Two mapping tables had new values:

  • Codes d'actes pour les médecins omnipracticiens (There appear to be less rows in the new)
  • Codes d'actes pour les médecins spécialistes (The appear to more rows in the new)

Only one data table has a different structure in the new data:

  • RAMQ_I_IQA2_SERV_REND_ACTE now has two more columns:
    • cod_modif_remu_acte - Appears to be a modifier of some kind for the amount paid. (Links to modificateurs)
    • cod_sect_ativ_etab - Appears to be the place where the acte took place (Links to secteur d'activité)

Interestingly, all of the data tables have slightly different byte sizes. It would be interesting to see what differences there are in the data.

Differences in the Fourth Request

The fourth dataset was sent on a set of 3 DVDs directly from RAMQ. Essentially, Jimmy identified a problem in which private clinics were not being included. This greatly changed the dataset, because people were only sampled if they had at least one visit to a private clinic for ILI. By including private clinics, the number of people and the number of visits per person rose dramatically.

The increase in the number of visits was so large (about 10 times) that is was somewhat unbelievable. I (Aman) verified that these results were probably correct, which I documented on a separate page.

Metadata

The metadata for Etude 1 is available here. It's all in French, and isn't the best. The table names in the database should be very similar if not exactly the same as the document.

The metadata for the address file is available here. Again, all in French, and not the best.

Deaths

Another dataset was received by CD, sometime late November in 2009, from the Institut de la statistique du Québec, containing the information on deaths. A document was received with the data. The CD is labeled:

DS 81 ISQ, David Buckeridge
CAI réf: 07 15 25
Institut de la statistique du Québec
27 Octobre 2009

The data arrived on a CD that was PGP encrypted. The password is located on this page only accessible to Aman.

After using the password to decrypt the file, it output a SAS data file, called Resultatdb.

I have described these data in full on the Etude 1 Deaths page.


CategoryData

Etude 1 (last edited 2012-04-30 12:44:02 by AmanVerma)