Analysis of Subjects with Multiple Encounters

[has mas]

Hello, new to the forum here. I'm a physician with a limited but thus far functional understanding of stata. I'm running into a problem with the following data set.

The overall analysis is looking at the effect of a drug recall on patients receiving a particular eye injection. The time of the recall is defined as t0. Each patient's eye in the data set has a unique code. Each encounter (i.e. office visit) at which time a patient would have received an injection is its own row (detailing the drug received). Therefore, if a patient's eye had multiple injections, then there will be multiple rows with the same code for that eye. For the rows, e-1 is the encounter before t0, e-2 the penultimate before t0, etc. e1, e2, etc are for encounters after t0. I'm having an issue figuring out how to do the analysis below in this nested layout of my data.

What would be the best commands for the following:
- Analyzing how many injections were received by each patient prior to t0
- How many patients switched drug from e-1 to e1 (before and after the recall)
- Categorizing the diagnostic indication for each eye (i.e. not by row but rather by eye).

I have tried searching in the manual and online, but I have hit a will. Hope my questions are clear- if not happy to clarify. Again, I'm primarily a clinical physician.

Appreciate any help.

[Clyde Schechter]

It will surprise me if anybody can answer this question based on the information presented. Even the most careful explanation in words of a data set is rarely adequate to support development of code.

I recommend that you post back showing example data from your data set. Use the -dataex- command to do that. If you are running version 18, 17, 16 or a fully updated version 15.1 or 14.2, -dataex- is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

[has mas]

Thanks for the suggestion. Had an issue with dataex (possibly because of the size of the data set). I have instead posted a condensed sample (removed many variables to keep anonymous/make it easier to read) from the Excel data sheet I'm working from.

PersonID=Patient level
SubjectID: Eye level data
Encounter number - all relative to T0 which is the recall date we are interested in
Injection number - again relative to T0, the recall date
Drug - Type of injection administered (names will be coded for the analysis, ignore empty cells, just encounters where a drug was not administered)

I am interested in data at the eye level (i.e. subjectID). Each eye may have several rows because it may have received multiple injections at multiple encounters

Again, looking to analyze for example the following questions:
- Analyzing how many injections were received by each eye (i.e. subjectID)
- How many patients switched drug from e-1 to e1 (before and after the recall)
- Comparing logmar by SubjectID from e-1 to e1

PersonID	SubjectID	SubjectEye	Days Before or After Recall	Encounter Number (Relative to T0)	Injection Number (Relative to T0)	Interval	Drug	logMAR
0A41087C0DFCA6A83879415F72542663	B65D351611CFAB93BDEFE053B732C6EB	Right	-118	-2	-2			0.87506126
0A41087C0DFCA6A83879415F72542663	B65D351611CFAB93BDEFE053B732C6EB	Right	-97	-1	-1		AVASTIN	1.09691001
0AA0FB142EDE523F30BAF3C7C1ADC27D	B1209C5C5659D8C7F1B6B9958C91B2C9	Right	-3	-1	-1			1.8
0AA0FB142EDE523F30BAF3C7C1ADC27D	B1209C5C5659D8C7F1B6B9958C91B2C9	Right	2	1	1	-35	AVASTIN	0.69897
0AA0FB142EDE523F30BAF3C7C1ADC27D	B1209C5C5659D8C7F1B6B9958C91B2C9	Right	37	2	2		AVASTIN	0.39794001
0AB4133CFF9D354B141976CEEAB7F55D	EC841715095AA5E6408529931FA34C4E	Right	-47	-2	-2	-28	AVASTIN	1.24303805
0AB4133CFF9D354B141976CEEAB7F55D	EC841715095AA5E6408529931FA34C4E	Right	-19	-1	-1	-35	AVASTIN	1.24303805
0AB4133CFF9D354B141976CEEAB7F55D	EC841715095AA5E6408529931FA34C4E	Right	15	1	1	-28	AVASTIN	1
0AB4133CFF9D354B141976CEEAB7F55D	EC841715095AA5E6408529931FA34C4E	Right	43	2	2		AVASTIN
0AB670293AEAB6109E8F864229458D97	15531923EFC1DD1BE4EFCF4DACE65245	Left	-171	-2	-2	-70	AVASTIN	0.39794001
0AB670293AEAB6109E8F864229458D97	15531923EFC1DD1BE4EFCF4DACE65245	Left	-101	-1	-1		AVASTIN	0.30103
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	-178	-6	-6	-29	AVASTIN	0.47712125
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	-149	-5	-5	-35	AVASTIN	0.54406804
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	-114	-4	-4	-175	AVASTIN	0.69897
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	-86	-3	-3			0.60205999
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	-44	-2	-2			0.54406804
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	-2	-1	-1			0.60205999
0AE1E0F30C33DE7FB959EC88E3D1E03E	615350DCA8DE7EB37166B2A59B8B3431	Left	60	1	1		EYLEA	0.47712125
0AE79B536650BDBA52C9EC3DB0020BD5	74742115B85F01B428907866D6A0FD77	Left	-152	-3	-3	-70	AVASTIN	1
0AE79B536650BDBA52C9EC3DB0020BD5	74742115B85F01B428907866D6A0FD77	Left	-82	-2	-2	-57	AVASTIN	0.87506126
0AE79B536650BDBA52C9EC3DB0020BD5	74742115B85F01B428907866D6A0FD77	Left	-25	-1	-1	-62	AVASTIN	0.87506126
0AE79B536650BDBA52C9EC3DB0020BD5	74742115B85F01B428907866D6A0FD77	Left	36	1	1		AVASTIN	0.87506126
0AF9CAF31DD5402B89280E4B4FE7207C	55162476A1186A566191801356005517	Left	-146	-2	-2	-99	AVASTIN	2.2
0AF9CAF31DD5402B89280E4B4FE7207C	55162476A1186A566191801356005517	Left	-47	-1	-1	-97	AVASTIN	2.2
0AF9CAF31DD5402B89280E4B4FE7207C	55162476A1186A566191801356005517	Left	49	1	1		AVASTIN	2.2
0BDCB30D962016C2C26D9E045E90FE08	E55B59DBD12FC2DD95A0CEF3DFF8D887	Left	-132	-1	-1		AVASTIN	0.60205999
0BDD1E0D9E4B781757B6FF8919C9E4E3	EDEDC591E14F484066953BCFF3583608	Right	-145	-2	-2	-104	AVASTIN	0.39794001
0BDD1E0D9E4B781757B6FF8919C9E4E3	EDEDC591E14F484066953BCFF3583608	Right	-41	-1	-1		AVASTIN	0.30103
0C5E91426D746B15855198809122D903	327D66D9D4E9CC0B006EF0DC30503518	Left	-132	-2	-2	-84	AVASTIN	0
0C5E91426D746B15855198809122D903	327D66D9D4E9CC0B006EF0DC30503518	Left	-48	-1	-1	-84	AVASTIN	0.09691001
0C5E91426D746B15855198809122D903	327D66D9D4E9CC0B006EF0DC30503518	Left	35	1	1		AVASTIN	0
0C8C8288739FF11D1DDD3F76A42ECC91	9239570C94B2EF82166F49EA04EB66D0	Right	-152	-2	-2	-84	AVASTIN	0.87506126
0C8C8288739FF11D1DDD3F76A42ECC91	9239570C94B2EF82166F49EA04EB66D0	Right	-68	-1	-1	-85	AVASTIN	0.87506126
0C8C8288739FF11D1DDD3F76A42ECC91	9239570C94B2EF82166F49EA04EB66D0	Right	16	1	1		CIMERLI	0.87506126
0CC7C1D71DCFCF107BED3515A8555D47	EAF654BA7C64E0E1FE2D2C130F670DCE	Left	-139	-2	-3		AVASTIN	0.47712125
0CC7C1D71DCFCF107BED3515A8555D47	EAF654BA7C64E0E1FE2D2C130F670DCE	Left	-41	-1	-2			0.54406804
0CC7C1D71DCFCF107BED3515A8555D47	EAF654BA7C64E0E1FE2D2C130F670DCE	Left	21	1	-1			0.47712125
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-178	-8	-9			0.69897
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-152	-7	-8			0.87506126
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-138	-6	-7			0.87506126
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-124	-5	-6		AVASTIN	0.87506126
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-96	-4	-5			0.87506126
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-61	-3	-4			0.69897
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-13	-2	-3			0.69897
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	-5	-1	-2			0.69897
0CD52628AC925ED60C7A0CBFAEC2DF1A	95425F119B5ACDD321E44B1EB1EA0F19	Left	29	1	-1			0.69897
0CD53908CB92CCB9FD7B92640B40F9F9	48B9FAE42A5588D4F52CD9CB24074C0F	Right	-114	-2	-2	-101	AVASTIN	0.39794001
0CD53908CB92CCB9FD7B92640B40F9F9	48B9FAE42A5588D4F52CD9CB24074C0F	Right	-13	-1	-1		AVASTIN	0.17609126
0CD69076FD59F188C8E994ED1E77343E	289B557518D16578420349A2DA2C2782	Left	-114	-2	-2	-98	AVASTIN	0.17609126
0CD69076FD59F188C8E994ED1E77343E	289B557518D16578420349A2DA2C2782	Left	-16	-1	-1		AVASTIN	0.09691001
0CFE3D6662B46E2E861F7B806C971886	83E9721C0509C2E6BDCF570ECD2A27F1	Left	-137	-2	-2			0.39794001
0CFE3D6662B46E2E861F7B806C971886	83E9721C0509C2E6BDCF570ECD2A27F1	Left	-101	-1	-1	-112	AVASTIN	2.1
0CFE3D6662B46E2E861F7B806C971886	83E9721C0509C2E6BDCF570ECD2A27F1	Left	10	1	1		EYLEA	2.2
00D1BB409256E80B55C477976BD600B8	CC720FF752645836C5FEB5C78145A68F	Right	-12	-2	-2			0.69897
00D1BB409256E80B55C477976BD600B8	CC720FF752645836C5FEB5C78145A68F	Right	-5	-1	-1	-28	AVASTIN	1
00D1BB409256E80B55C477976BD600B8	CC720FF752645836C5FEB5C78145A68F	Right	22	1	1		AVASTIN	0.47712125

[Mike Lacy]

Your data example as posted is quite messy to work with, and won't much help people help you for reasons too lengthy to explain here. It's quite possible to use -dataex- to display example data for a limited number of observations and variables, which I'm guessing (?) will help you. Try the following to produce a workable -dataex- example that you can post here:

encode PersonID, generate(pid)
dataex pid  PersonID  SubjectID SubjectEye etc. if inlist(PersonID,1, 2, 3)

The preceding will enable you to get a -dataex- listing for the 3 persons and your chosen selection of variables. The individuals chosen will be the first three in alphabetical order of PersonID, which may not be optimal to reveal the details of your data, but would be a good start, and will be much more readable. Where I show "etc." you need to include a list, separated by spaces, of the variable names of interest.
(You might benefit by refreshing your memory by looking again at some of the examples shown by -help dataex-) This should produce a listing you can post back here. If this doesn't work for you, ask for more help with -dataex-, explaining just what your problem with it was.