Youtube gives you help how to perform repeated measures on SPSS.

You cannot compare 62 items with 34 items in case you cannot delete 28 items from those 62.  When you plan your research you can decide beforehand how to compute your data. If you decide to do repeated measures for 3 times with 34 items in each it may be OK.

For the interventions you can use ANOVA for possible significanses between independent groups.

You can compute a regression analysis where your total scores on clinical assessment measure is your dependent variable and then analyze it by your independent variables: Counseling only, medication only, and counseling + medication.

Ok. My participants get a standardized score for each measured assessment. So, I should remove the 28 items from the baseline measure so that it is consistent with the subsequent measures. Recalc the baseline standardized score and then move forward w/ the regression analysis.

That is right! Good luck!

In regression model you can include scores at baseline as an independent measures to control for differences in the beginning. You'll end up three analyses one for each follow up.

This I do not understand? What would the baseline score explain?

I'll echo the comment about ensuring that all your assessments are on the same scale. In my opinion, there are better ways to analyze this data than running several independent regressions. Repeated measures ANOVA is certainly a possibility, though this doesn't allow you to model variability in your intercepts and slopes (if there is any). I would suggest you look into using multilevel modeling, with assessments at level 1 (perhaps predicted by time since since baseline assessment) and condition (dummy/contrast coded), number of sessions, and any other between-subjects covariates at level 2. Willet & Singers ALDA book should give you a good start. Best of luck! 

Men like to echo female answers. However if you do an independent variable of e.g. baseline as suggested look for multicollinearity as "a statistical phenomenon in which two or more predictor variables in a multiple regression model are highly correlated, meaning that one can be linearly predicted from the others with a non-trivial degree of accuracy. In this situation the coefficient estimates of the multiple regression may change erratically in response to small changes in the model or the data. Multicollinearity does not reduce the predictive power or reliability of the model as a whole, at least within the sample data themselves; it only affects calculations regarding individual predictors. That is, a multiple regression model with correlated predictors can indicate how well the entire bundle of predictors predicts the outcome variable, but it may not give valid results about any individual predictor, or about which predictors are redundant with respect to others." Wikipedia

The trouble with YOuTube is that it tells you how to do something right that may be the wrong thing to do.

Repeated measures ANOVA makes assumptions about your data that are unlikely to be true anywhere outside a black hole. 

You are better off using robust variance estimates to account for clustering of data within participants. This allows you to use all available data at all time points. 

The main danger with all missing data is that the participants in whom data are missing may be different to those who continue. 

And be careful of Wikipedia (and even more so of quotes from Wikipedia). The problem with collinearity is that the resulting analysis may test a model that makes little real-life sense. If two variables are strongly correlated, then it may make no sense to ask what happens when you hold one of them constant. The effect of holding one constant may be that there is almost no remaining variation in the other. So you need to specify sensible hypotheses to test. 

Just to clarify the recommendation to adjust for baseline covariates, this increases the statistical efficiency of the analysis. It does so essentially by reducing the error variance in the model, which is used as a basis for calculating the significance associated with a particular regression term.

See this draft policy document from the European Medicines Agency. It states, in part: 

In case of an ordinary linear model, if a baseline value of a continuous outcome measure is available, then this should usually be included as a covariate. This applies whether the primary outcome variable is defined as the ‘raw outcome’ or as the ‘change from baseline’.

http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2013/06/WC500144946.pdf

Every participant should have the same number of observations or assessment to draw substantive conclusions in a macro sense. The main concerns I have, assuming the assessment measures are valid and reliable, are subjective bias, demand characteristics, and maturation. Then also, are the measures taken before or after the treatment sessions in each testing interval. I prefer the latter for the previous concerns. Check out xpainstress.com for a mental health services questionnaire in use with chronic pain/stress of single subject/group repeated measures over time.

Thank you all.

Is the 62 item baseline measure valid or standardized? What is the DV clinical assessment measure?

I would recommend using multilevel modeling to account for the nested nature of your data (observations nested within individuals). One of the benefits of MLM is that you don't need the same number of observations for each case.